JP4638824B2 - Production / distribution plan creation apparatus, production / distribution plan creation method, distribution control apparatus, distribution control method, program, and computer-readable recording medium - Google Patents

Description

  The present invention relates to a production / distribution plan creation device in a production process / conveyance in which a plurality of products are processed or conveyed in a plurality of different processes or conveyance routes, and each product can select a different multi-step route or conveyance route. The present invention relates to a distribution plan creation method, a distribution control device, a distribution control method, a program, and a computer-readable recording medium.

  2. Description of the Related Art Conventionally, in manufacturing processes and transportation in many industries such as steel, a plurality of products are processed in a plurality of different processes, and each product is manufactured and transported under a condition that a plurality of different process paths can be selected.

  For example, in the ironworks, the process of transferring the raw material from the raw material yard to the raw material storage tank, multiple brands are stacked for multiple tanks where different brands are introduced and discharged at different cutting speeds. In addition to selecting a pile of raw materials that matches the brand in the tank from a plurality of yards, an appropriate reclaimer is selected from a plurality of reclaimers that can be used.

  Then, the selected reclaimer performs a cutting process from the raw material pile, and the cut iron ore selects an appropriate belt conveyor series from among a plurality of belt conveyor series that can be transported, transports it to the raw material storage tank, and stores it in the raw material storage tank. Enter the appropriate tank volume from the appropriate start time to the end time.

  Furthermore, a plurality of brands are stacked in a plurality of yards, and the same brand may be stacked on a plurality of raw material piles.

  When planning the production of raw materials for slag from the yard to the raw material storage tank, it is necessary to take into account the operational restrictions of the yard, reclaimer, belt conveyor series, raw material storage tank, and restrictions due to the raw material logistics process. There is.

  That is, in the raw material factory, the raw material storage tank must not be out of stock for the purpose of stabilizing the operation of the blast furnace operation and the sintering factory. For this reason, it is necessary to constantly monitor the inventory changes of a large number of raw material storage tanks, and always take care.

  In addition, when the stock level of the raw material storage tank falls below a certain level, the coarse iron ore that has accumulated in the tank flows out at once, causing the iron ore to lose its grain size stability, or in the case of sintered ore. There is a concern that pulverization may occur due to a large drop distance when the tank is placed.

  In order to prevent these, the stock of raw material storage tanks is required to be highly stable. However, in this case, in order to further reduce the work load, efficient use of equipment, that is, reduction in operating rate is also required.

  Furthermore, since there are multiple process paths when entering the same raw material storage tank as well as different process paths for each brand entering the raw material storage tank, it is necessary to judge the usage status of the equipment and select an appropriate process path. is there. In addition, it is necessary to take into consideration that the processing time in each process and each equipment is different.

  Under such various constraints, if you use one reclaimer from one raw material pile, convey it by one belt conveyor system, and enter into one raw material storage tank, simply stock the raw material storage tank. If the level is low, you can carry it.

  However, under the operating conditions where multiple brands as described above are stacked in multiple yards and the same brand is stacked in different raw material piles at multiple locations, the overall production efficiency In order to improve the raw material storage tank, it is necessary to make a raw material yard operation plan for which raw material storage tank should be used in what order and which reclaimer and belt conveyor system will be used, and from when to when the tank is put into operation. In order to stabilize the operation of the factory, it is possible to secure inventory, stabilize the iron ore particle size, and prevent the sinter ore from being pulverized.

  In general, in a manufacturing process in which a plurality of products are processed and manufactured in a plurality of different processes, production / distribution planning in the process or automation of distribution control is desired. Conventionally, as a technique for this automation, a discrete event simulator or the like by various methods has been proposed. For example, as disclosed in “Production Plan Evaluation Method and System” of Patent Document 1, on a simulator simulating a factory built on a computer, information obtained from the same interface as the actual device is used. By predicting the operation of the product, performing virtual production at a faster speed than the actual equipment based on the operation prediction, and presenting a highly accurate index using the virtual production process and results, the production plan It is a technique that enables evaluation and selection.

JP 2002-366219 A

  As disclosed in the “Production Plan Evaluation Method and System” described in Patent Document 1 above, the method of creating a production / distribution plan using a conventional discrete event simulator requires a satisfactory result. (1) It was necessary to perform simulations while changing the conditions in various ways, and repeatedly evaluate the results. Therefore, (2) a large-scale factory has a problem that it takes a lot of time to create a production / distribution plan. In addition, (3) in order to obtain a highly accurate production / distribution plan, there is a problem that simulation rules must be set in detail.

  Conventionally, simulators using various methods have been proposed as techniques for production / distribution plan creation or automation of distribution control.

  However, as in the above operation example, there are many limitations, and a method that is particularly suitable for executing an operation plan or distribution control for selecting an appropriate process route while keeping the constraints has not been proposed. It was. In particular, in order to withstand use in actual operations, production / distribution planning or distribution control that can be flexibly optimized in accordance with the planner's intention at high speed or with the accuracy required by the planner is required. However, a method particularly suitable for the above request has not been proposed.

  Therefore, the present invention is a production / distribution process in which a plurality of products are processed in a plurality of different processes, and each product can select a plurality of different process paths (even if different facilities are used in the same process). It is intended to be able to flexibly optimize production / distribution plans or distribution instructions that require arbitrary time accuracy at high speed or in detail with the accuracy required by the planner, in accordance with the planner's intention. .

The production / distribution plan creation device of the present invention is a production / distribution plan creation device for creating a production / distribution plan in a manufacturing process / conveyance in which a plurality of products are processed or conveyed by different plural process routes or conveyance routes.・ Suitable for simulation process that simulates the manufacturing process and transport that expresses the logistics state and logistics constraints of transportation, and for the instruction calculation period (macro instruction calculation period) that is set in advance from the start date and time of the production and logistics planning. A macro planning device that calculates a logistics instruction (macro logistics instruction) for the simulator by performing a process (macro optimization processing), and a macro logistics instruction obtained by the macro planning device, from the planning start date and time Optimization process for instruction calculation period (micro instruction calculation period) shorter than the preset macro instruction calculation period A production / distribution plan creation device including a micro-planning device that performs a micro-optimization process and calculates a logistics instruction (micro-distribution instruction) for the simulator. (A) First, from the planning start date and time, the macro Macro logistics instructions for the macro instruction calculation period are created by the planning device. (B) Given the macro logistics instructions, the logistics instruction is calculated for the micro instruction calculation period by the micro optimization process by the micro planning device. The simulation is performed for a preset simulation period, (c) the simulation result is confirmed as a production / distribution plan for a preset micro plan determination period, and (d) is determined. The date and time immediately after is set as a new planning start date and time, and the production / logistics plan that has already been confirmed is given. A new planning start date and time that sets a series of processes to execute a simulation based on the new logistics instruction obtained by the micro optimization process and to finalize the production and logistics plan for the new micro plan decision period (E) When the production / distribution plan for the macro plan confirmation period is confirmed by the above-described series of processes, a new planning start date and time is set, and the already confirmed production / distribution plan is given. A production / distribution plan for a desired plan creation period is created by performing a series of processes including the steps (a) to (e).
The production / distribution plan creation method of the present invention is a production / distribution plan creation method for creating a production / distribution plan in a manufacturing process / conveyance in which a plurality of products are processed or conveyed by different multiple process routes or conveyance routes, The simulation process that simulates the manufacturing process / convey that expresses the logistics state and logistics constraints of the manufacturing process / conveyance, and the target period (macro instruction calculation period) that is set in advance from the production / distribution planning start date and time A macro mathematical formula model construction step for constructing a mathematical formula model (macro mathematical formula model) composed of the logistics state and logistics constraints of the manufacturing process / conveyance based on preset time accuracy (macro time accuracy), and the macro An evaluation function (macro evaluation function) for evaluating the mathematical model is set, and the macro mathematical model, the macro evaluation function, Macro instruction calculation period that is set in advance from the planning start date and time given the macro formula model optimization process that uses the macro formula model optimization process and the logistics instruction (macro logistics instruction) obtained by the macro formula model optimization process For a shorter target period (micro instruction calculation period), based on a preset accuracy (micro time accuracy), a mathematical model (micro mathematical model) consisting of the above-mentioned manufacturing process / conveyance logistics state and logistics constraints A micro mathematical model construction process to be constructed and an evaluation function (micro evaluation function) for evaluating the micro mathematical model are set, and an optimization process is performed using the micro mathematical model and the micro evaluation function. A micro mathematical model optimization process for calculating logistics instructions for the simulator, (a) First, from the start date of planning A macro formula instruction for a macro instruction calculation period is created by a black formula model construction process, a macro evaluation function setting process, and a macro formula model optimization process, and (b) given the macro logistics instruction as a given, the micro formula model The distribution instruction is calculated for the micro instruction calculation period by the optimization process, and the simulation is executed in the simulation process for the preset simulation period. (C) The preset plan determination period (micro plan determination period) ) Confirm the result of the above simulation process as a production / distribution plan (production / distribution plan confirmation process), and (d) set the date and time immediately after the confirmation as a new planning start date and time. Based on the logistics plan, a new micro mathematical model is built, and the micro optimization process for the micro mathematical model A simulation is executed based on the new logistics instruction obtained by the above, and a series of processes for finalizing the production / logistics plan for the new microplan decision period is performed for each new planning start date and time to be set sequentially (micro Update step), (e) a macro execution determination step for determining whether a preset plan determination period (macro plan determination period) has been determined by the micro update step, and further, a macro plan determination period is determined. When setting a new planning start date and time, given the production / distribution plan already determined by the production / distribution plan determination step, a series of steps consisting of the steps (a) to (e), A production / distribution plan for a desired plan creation period is created.
The physical distribution control apparatus of the present invention is a physical distribution control apparatus for controlling physical distribution in a manufacturing process / conveyance in which a plurality of products are processed or conveyed by different plural process paths or conveying paths, and the distribution state and physical distribution of the manufacturing process / conveyance Optimization process (macro optimization process) for the simulator that simulates the manufacturing process / convey expressing the constraints and the instruction calculation period (macro instruction calculation period) set in advance from the start date of the production / distribution plan. ) To calculate the logistics instruction (macro logistics instruction) for the simulator, and the macro instruction calculation set in advance from the planning start date and time given the macro logistics instruction obtained by the macro planning apparatus. Perform an optimization process (micro optimization process) for an instruction calculation period (micro instruction calculation period) shorter than the period. A logistics control device comprising a micro-planning device that calculates a logistics instruction (micro-logistics instruction) for a simulator, and (a) first, macro logistics for a macro-instruction calculation period from the planning start date and time by the macro-planning device. (B) Given the macro logistics instruction, (b) By using the micro-optimization process by the micro-planning device, the logistics instruction is calculated for the micro-instruction calculation period, given to the simulator, and set in advance. (C) The simulation result is confirmed as a production / distribution plan for a predetermined micro plan determination period, and (d) the date and time immediately after the determination is set as a new planning start date and time. Newly obtained by micro optimization processing, given the production / logistics plan that has already been established. A series of processes for executing a simulation based on a flow instruction and finalizing a production / distribution plan for a new micro-plan finalization period is performed for each new planning start date and time set sequentially. (E) By the above series of processes When the production / logistics plan for the macro plan confirmation period is finalized, a new planning start date and time is set, and the production / logistics plan that has already been determined is given, and the processes (a) to (e) are performed. A production / distribution plan for a desired plan creation period is created by performing a series of processing, and based on the created production / distribution plan, the production process and the logistics in the process to be transported are controlled. It is characterized by.
The physical distribution control method of the present invention is a physical distribution control method for controlling physical distribution in a manufacturing process / conveyance in which a plurality of products are processed or conveyed by different multi-step paths or conveyance paths, Preliminary time accuracy (macro instruction calculation period) for a simulation process that simulates a manufacturing process / convey that expresses logistics constraints, and a target period (macro instruction calculation period) that is set in advance from the start date of production / distribution planning. Macro mathematical model construction process for constructing a mathematical model (macro mathematical model) composed of the logistics state and logistics constraints of the manufacturing process / transport based on the macro time accuracy), and an evaluation function for evaluating the macro mathematical model (Macro Evaluation Function) is set, and a macro that performs optimization processing using the above macro mathematical model and the above macro evaluation function Given the formula model optimization process and the logistics instruction (macro logistics instruction) obtained by the macro formula model optimization process, the target period (microinstruction calculation is shorter than the macro instruction calculation period preset from the planning start date and time) A micro mathematical model construction process for constructing a mathematical model (micro mathematical model) composed of the logistics state and logistics constraints of the manufacturing process / transport based on a preset precision (micro time precision) And setting an evaluation function (micro evaluation function) for evaluating the micro mathematical model, performing an optimization process using the micro mathematical model and the micro evaluation function, and calculating a logistics instruction for the simulator. (A) First, from the planning start date and time, the macro mathematical model construction process, A macro logistics instruction for the macro instruction calculation period is created by the evaluation function setting process and the macro mathematical model optimization process, and (b) the micro instruction is calculated by the micro mathematical model optimization process given the macro logistics instruction. The distribution instruction is calculated for the period, and the simulation is executed for the preset simulation period in the simulation process, and (c) the simulation process is performed for the preset plan decision period (micro plan decision period). The result is confirmed as a production / logistics plan (production / logistics plan decision process), and (d) the date and time immediately after the decision is set as a new planning start date and time. Build a micro mathematical model and create a new logistics instruction based on the micro optimization process for the micro mathematical model. (E) an update process (micro update process) for executing a simulation and performing a series of processes for determining a production / distribution plan for a new micro plan determination period for each new planning start date and time, A macro execution determination step for determining whether or not a predetermined plan determination period (macro plan determination period) has been determined by the micro update process, and when the macro plan determination period is determined, a new planning start date and time The production for the desired plan creation period is made by a series of steps consisting of the steps (a) to (e) given the production / distribution plan already determined by the production / distribution plan determination step.・ Processing to create a logistics plan and controlling the logistics in the manufacturing process and the process to be transported based on the production and logistics plan created above And wherein the Rukoto.
According to a first aspect of the program of the present invention, the production / distribution plan creation method is executed by a computer.
According to a second aspect of the program of the present invention, the distribution control method is executed by a computer.
The computer-readable recording medium of the present invention is characterized in that the program is recorded.

  According to the present invention, in a process that can process a plurality of products in a plurality of different processes, and that each product can select a different multi-process path (considering a different process path even when using different equipment in the same process), It is possible to flexibly optimize production / distribution plans or distribution instructions that require arbitrary time accuracy at high speed or in detail with the accuracy required by the planner and in accordance with the planner's intention.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, an embodiment of a production / distribution plan creation apparatus, a production / distribution plan creation method, a distribution control apparatus, a distribution control method, a program, and a computer-readable recording medium according to the present invention will be described in detail with reference to the accompanying drawings. To do.

First, the basic configuration of the production / distribution plan creation apparatus of this embodiment will be described with reference to FIG.
FIG. 9 is a block diagram illustrating an example of the configuration of a production / distribution plan creation apparatus to which the present invention is applied. As shown in FIG. 9, the production / distribution plan creation device 21 of this embodiment includes a simulator 910, a macro planning unit 920, and a micro planning unit 930. The simulator 910 is a large simulator that simulates a factory or the like.

  In this embodiment, product acceptance plan, product shipment plan, inventory plan, equipment use plan, equipment repair plan, equipment capacity, process status, equipment status, inventory status, equipment operation / failure status, and operation in manufacturing process / conveyance Based on the input data 950 representing all or part of the operation preconditions from the operator, the macro time calculation accuracy set in advance for the macro instruction calculation period set in advance from the production start date and time of the production / distribution plan Based on this, optimization calculation is performed by the macro planning unit 920 so as to satisfy the relationship and constraints of the work groups associated with the processing of the product, the moving body, and the equipment, and the macro logistics instruction for the micro planning unit is calculated. .

  Based on the macro logistics instruction and the micro data calculation period set in advance from the production / distribution planning start date and time based on the macro logistics instruction and the input data 950, the product, The micro-planning unit 930 performs an optimization calculation so as to satisfy the relationship and constraints of the work group accompanying the processing of the moving body and the facility, and calculates the physical distribution instruction for the simulator 910.

  Therefore, according to the production / distribution plan creation apparatus of the present embodiment, the distribution instruction is not performed based on a predetermined rule as in the past, but is performed by the macro planning unit 920 and the micro planning unit 930. A physical distribution instruction based on the result of the optimal calculation can be output to the simulator 910. As a result, it is possible to reliably issue an optimum physical distribution instruction according to the event at that time.

  In addition, when a new event occurs, a calculation instruction for causing the simulator 910 to perform a calculation to the macro planning unit 920 and the micro planning unit 930 is output. When the calculation instruction is given from the simulator 910, optimization calculation is executed using the macro planning unit 920 and the micro planning unit 930. As described above, an optimal production / logistics production / logistics plan can be created by executing a detailed simulation that links the simulator 910, the macro planning unit 920, and the microplanning unit 930 for each event once. it can.

  That is, the simulation performed in the present embodiment is not based on a predetermined rule as in the prior art, but is performed based on the result of the optimal calculation, so only one simulation is performed. It is possible to reliably obtain a theoretically optimal solution, and it is not necessary to repeatedly evaluate the simulation result by evaluating the simulation result as in the past, and the simulation result 940 can be generated quickly and with high accuracy. it can. Therefore, even if the production / distribution plan is created on a large scale, it can be sufficiently created within the practical time. The simulation result 940 obtained as described above is output as a production / distribution plan.

  Further, even if the simulator 910 is very large, or there are very many restrictions and complicated, among the distribution conditions and distribution restrictions described in the simulator 910, the production / distribution plan creation is greatly affected. By taking only the important part into the macro plan part and the micro plan part, the scale of the simulator 910 can be set within an appropriate range and the optimization calculation can be performed within a practical time. .

  Since the simulator 910 can describe all the distribution conditions and distribution restrictions to be considered, it is guaranteed that the production / distribution plan created by performing one simulation is actually executable.

  As described above, in this embodiment, the simulator 910, the macro planning unit 920, and the micro planning unit 930 are linked to create the logistics production / distribution plan. (1) The simulation is not repeated. It is possible to create a production / logistics plan. In addition, (2) calculation time can be shortened by incorporating only important parts that have a large impact on production / distribution planning into the macro planning and micro planning departments, and (3) large-scale problems. Can be solved.

  In addition, whenever an event requiring a physical distribution instruction occurs, the physical state and physical distribution constraint information of the simulator 910 is detected, and the macro planning unit 920, the microscopic unit 920, and the micro evaluation unit are based on the detected information and a predetermined evaluation index. Since the planning unit 930 calculates the optimal logistics instruction by the optimization method and creates a production / distribution plan by performing a detailed simulation with the simulator 910 based on the calculation result, (4) increase production / distribution planning accuracy (5) It is possible to create a production / distribution plan whose feasibility is verified.

  In addition, the macro planning department and the micro planning department have two stages of optimization processing, and the macro planning department performs optimization considering long-term production and logistics. Since it is possible to realize detailed optimization in a short period of time upon receiving a logistics instruction, (6) Production / Logistics that takes into account the long-term impact and has sufficient optimization performance even in a short period of time A plan can be created.

  In the production / distribution plan creation apparatus of this embodiment, raw materials are used as examples of manufacturing processes in which a plurality of products are processed or transported in a plurality of different processes or transport paths, and each product can select a different process path or transport path. A case where a production plan or a distribution plan is created will be described using an example in which a transfer path from a yard to a plurality of raw material storage tanks can be selected.

  In the present embodiment, the production / distribution plan means a production or / and distribution plan. Similarly, the manufacturing process / transport means a manufacturing process or / and transport.

  And in order to stabilize the operation of the blast furnace / sinter factory under the constraints of raw material logistics such as yard loading brand, yard inventory change, iron ore / sinter cut amount, equipment layout, etc. In order to prevent pulverization of sintered ore and to stabilize the stock level, and to reduce the work load, we will deal with the optimization problem of the raw material yard entry plan that realized the efficient use of equipment.

  However, this is merely an example, and the production / distribution plan creation apparatus of the present embodiment processes a plurality of products in a plurality of different processes or transport routes or transports them by a plurality of moving bodies, and a plurality of processes in which each product is different. In the manufacturing process / conveyance in which a route or a conveyance route can be selected, the present invention can be applied when creating an operation plan in the target process while keeping many restrictions, and is particularly effective.

  In the operation plan here, first of all, we secure the stock to stabilize the operation of the blast furnace and sintering plant (prevention of out of stock), stabilize the iron ore particle size, and prevent the sinter ore from being pulverized Therefore, the purpose is to realize the high level of inventory level and the efficient use of equipment to reduce the work load.

  In the manufacturing process according to the present embodiment, the amount to be filled is different because the amount of cut out is different for each raw material storage tank and the tank entry conditions such as the stock level of the corresponding raw material storage tank are different at the time of starting the tank. It is necessary to determine that the raw material storage tank level is highly stable depending on the situation.

  In addition, as shown in FIG. 1 which is a schematic diagram of a raw material yard that is an implementation target of the production / distribution plan creation apparatus of the present embodiment, a plurality of reclaimers and belts are used for conveyance to a raw material storage tank to be entered. A set of conveyor series can be selected.

  In addition, since the above-mentioned plurality of reclaimers have different cutting-out capacities, and there are few reclaimers compared to the large number of raw material storage tanks, the reclaimers often come together and the degree of freedom in selecting the transport path is large. In the raw material yard manufacturing process which is one implementation target of the production / distribution plan creation apparatus of the form, it is necessary to appropriately select an appropriate reclaimer and a belt conveyor system and operate for an appropriate time.

  Under such constraints, when securing the stock of all raw material storage tanks and creating a raw material yard operation plan with a high level of inventory, the order of tank entry, tank start and end times, tank volume, reclaimer It is necessary to accurately determine not only the operation start time and reclaimer operation end time, but also the payout raw material mountain, yard, used reclaimer, conveyor belt conveyor series, and incoming tank raw material storage tank. In this case, it is necessary to have both high speed and accuracy that can withstand use in actual operation.

  FIG. 2 is a diagram showing the relationship between the configuration of the raw material yard entry plan creation measure using the production / distribution plan creation device according to the present embodiment and other systems.

  As shown in FIG. 2, a raw material yard incoming tank plan creation device is configured using an incoming tank plan creation unit (production / plan creation device) 21. When creating a raw material yard tank plan, first, the condition setting and take-in unit 20 operates the constraint conditions, capacity conditions, and preconditions, such as the yard layout, raw material storage tank cutout amount, etc., which are necessary for planning. The person takes in the data from the setting or the process control or vidicon.

  The tank entry plan creation unit 21 of the present embodiment obtains a tank entry plan for the raw material yard so as to satisfy these distribution restrictions, capacity conditions, and the like under various distribution restrictions set by the condition setting and taking-in part 20.

  That is, the order of tank entry, tank entry start / end time, start / end time of reclaimer operation, discharge raw material mountain / yard, used reclaimer, conveyor belt conveyor series, and input tank raw material storage tank are obtained.

  As will be described in detail below, the tank plan creation unit 21 uses macro and micro formulas using mathematical programming methods such as LP (linear programming), MIP (mixed integer programming), and QP (quadratic programming). By constructing a model, or by constructing a genetic model or a priority model without constructing a mathematical model using a meta-heuristic method such as tab search or genetic algorithm, or a combination of a meta-heuristic method and mathematical programming In this way, the processing order, processing time, raw material equipment to be used, and transport route are optimized from the raw material yard to the raw material storage tank.

  Raw material yard entry plan obtained by the production / distribution plan creation unit 21 (order of entry, entry / exit time, entry / exit time of reclaimer, and delivery material pile / yard, reclaimer used, belt conveyor series to be conveyed , Information on the incoming tank raw material storage tank) is given to the display unit 22, and is displayed in a form such as a Gantt chart format, raw material storage tank inventory transition graph format, or an incoming tank time list.

  The operator evaluation section 23 evaluates the obtained tank entry plan from various viewpoints (for example, inventory transition, continuous issue of the same brand in the reclaimer, etc.). Correct the order of tank entry, start / end time of tank entry, dumped raw material pile, reclaimer used, etc.

  In this case, the weight of the evaluation function or the evaluation index is changed as necessary, the target period / plan decision period for constructing the mathematical model, the gene model, and the priority model is changed, or all or It is equipped with means that can only fix the entry time, specify the pile of raw materials to be dispensed, and specify and fix the reclaimer to be used only for the designated treatment. Then, the tank entry plan creation unit 21 creates the tank entry plan again.

  Next, the structure of the said tub plan preparation part 21 and the detail of the process performed by the basin plan preparation part 21 are demonstrated.

  The tank entry plan creation unit 21 first has a long-term macro view, with a rough accuracy, under the setting conditions such as the yard layout, process route, tank entry brand, and distribution constraints, and the stock quantity and raw material delivery speed for each raw material storage tank. Therefore, the logistics constraints such as tank inventory transition for each raw material storage tank, process route selection, etc. are expressed as a mathematical model, avoidance of raw material tank stock out of stock, high stability of stock level of raw material storage tank and efficient use of equipment to reduce workload. Therefore, the tank destination, the process route and the approximate tank entry time that optimize the predetermined evaluation function set for the purpose are determined.

  Next, the tank destination and the process route are set as predetermined conditions set for the use of efficient equipment to avoid out of stock of the raw material storage tank, to stabilize the stock level of the raw material storage tank and to reduce the work load. In order to optimize the evaluation function, the order of tank entry, the start and end time of the tank, the amount of tank input, the start time of the reclaimer operation, and the end time of the reclaimer operation are determined in detail from a microscopic viewpoint, The starting and ending time of the tank, the amount of tanks, the start time of the reclaimer operation, and the end time of the reclaimer operation, as well as the discharge raw material mountain, the yard, the used reclaimer, the conveyor belt conveyor series, and the incoming tank raw material storage tank are determined. This process is performed to update the time of the target period to obtain a tank entry plan for a desired period.

  Here, in order to explain the details of the processing of the tank entry plan creation unit 21 described above, the raw material yard manufacturing process (conveyance) will be described with reference to a simplified diagram (FIG. 5). In the example of FIG. 5, the first yard has a pile of raw materials on which iron ore brands A, B, and C are stacked, and the second yard is loaded with brand B.

  The reclaimer No. 1 can be used to pay out the raw material pile in the first yard, and the reclaimer No. 2 can be used to pay out the raw material pile in the second yard.

  When the reclaimer No. 1 is used, the iron ore is conveyed by any one of the first belt conveyor series, the second belt conveyor series, the third belt conveyor series, and the fifth belt conveyor series.

  On the other hand, when the reclaimer No. 2 is used, the iron ore is conveyed by either the fourth belt conveyor series or the sixth belt conveyor series.

  The iron ore transported by the first belt conveyor series is transferred to the raw material storage tank 1, and the iron ore transported by the second belt conveyor series or the fourth belt conveyor series is transferred to the raw material storage tank 2 and the third belt conveyor. The iron ore conveyed in the series is conveyed to the raw material storage tank 3, and the iron ore conveyed in the fifth belt conveyor series or the sixth belt conveyor series is conveyed to the raw material storage tank 4, respectively.

  Brand A must be placed in the raw material storage tank 1, brand B in the raw material storage tank 2, brand C in the raw material storage tank 3, and brand B in the raw material storage tank 4. Here, the brand to be paid out from the yard and the brand to be placed in the raw material storage tank must be the same brand.

-First embodiment-
As a first embodiment, a detailed configuration of the production / distribution plan creation device (incoming tank plan creation unit) 21 is an embodiment that uses a macro mathematical model and a micro mathematical model for the details of processing of the tank entry plan creation unit 21. FIG. 3 (FIG. 3), a flowchart (FIG. 4) showing the detailed processing contents of the production / distribution plan creation device (incoming tank plan creation unit) 21, and the production / distribution in the case of using the example of FIG. This will be described in detail with reference to the drawings (FIGS. 6 to 8) showing the details of the operation inside the plan creation device (the tank entry planning unit) 21. FIG.

Here, FIG. 3 is a diagram showing a detailed configuration of the production / distribution plan creation apparatus according to the present embodiment with respect to the basic configuration of the production / distribution plan creation apparatus according to the present embodiment described with reference to FIG. .
Here, the simulation process is step S411 in FIG. 4, the macro mathematical model construction process is step S407 in FIG. 4, the macro mathematical model optimization process is step S408 in FIG. 4, and the micro mathematical model construction process is step S409 in FIG. The formula model building process comprises step S410 in FIG. 4, the production / distribution plan finalizing process comprises step S412 in FIG. 4, the micro update process comprises step S415 in FIG. 4, and the macro execution determination process comprises steps S414 and S416 in FIG. The

  (1) Capture input data and set initial values and conditions (input data capture unit 301 in FIG. 3, step S401 in FIG. 4). Specifically, information necessary for this treatment (raw material acceptance plan, raw material yard plan, equipment repair plan, raw material yard status, tank inventory status, tank cut-out status, equipment operation / failure status, and information from operators All or part of the operational prerequisites are read online and the operator modifies them as necessary.

  (2) A production / distribution plan creation period is set (plan creation period setting unit 302 in FIG. 3, step S402 in FIG. 4). In this process, a period for creating a production / distribution plan is set. This creation period can be set as desired according to the planner's needs. Here, as an example, 4 days are planned.

  (3) The production / distribution plan creation time accuracy is set (plan creation time accuracy setting unit 303 in FIG. 3, step S403 in FIG. 4). In this process, macro time accuracy, micro time accuracy and simulation accuracy for creating a production / distribution plan are set. The macro time accuracy, micro time accuracy, and simulation accuracy can be set arbitrarily according to the needs of the planner.

  For example, the macro time accuracy is coarse, and after roughly determining the production / logistics plan from a long-term perspective, the production / logistics plan is determined in detail with finer micro-time accuracy based on the decision result. Therefore, it is possible to create an efficient plan with sufficient accuracy and a short time from a long-term perspective.

  As another example, sufficient accuracy and short time can be achieved by reducing the accuracy in the first half of the planning period that requires fine planning accuracy, and coarsening the accuracy in the second half of the sufficient planning period. Can create efficient plans.

  Here, as an example, when creating a macro formula model for the first time, the macro time accuracy is set to 1 hour, and when creating the second and subsequent macro formula models, the macro time accuracy is set to 2 hours. . In the micro time accuracy and simulation part, the accuracy is made minute throughout the planning period. This makes it possible to create a production / logistics plan with the accuracy that can be applied as it is in actual operation by building a mathematical model that can be solved accurately in a short time and performing simulation with the accuracy required in actual operation. To do.

  (4) A production / distribution instruction calculation period is set (instruction calculation period setting unit 304 in FIG. 3, step S404 in FIG. 4). In this process, a macro instruction calculation period and a micro instruction calculation period for creating a production / distribution plan are set. In the macro instruction calculation period and the micro instruction calculation period, any target period can be set individually as required by the planner.

  For example, the macro instruction calculation period covers a long period of time in order to create a production and logistics plan from a sufficiently long-term perspective, and enables micro-planning in a short time based on the results of the macro-planning. Therefore, by shortening the micro instruction calculation period, it is possible to create a plan that is sufficiently optimal, sufficiently accurate, and efficient in a short time from a long-term perspective.

  Here, as an example, the macro instruction calculation period is set to 3 days and the micro instruction calculation period is set to 1 day throughout the planning period, so that a sufficiently long range can be optimized and efficient planning can be made in a short time. I can do it.

  (5) A production / distribution plan decision period is set (plan decision period setting unit 305 in FIG. 3, step S405 in FIG. 4). In this process, a macro plan finalizing period and a micro plan finalizing period for finalizing the production / distribution plan are set. The macro plan finalization period and the micro plan finalization period can be arbitrarily set as required by the planner.

  For example, by shortening the plan decision period in the first half of the plan creation period that requires fine planning accuracy, and increasing the plan decision period in the second half of the plan creation period that is sufficient for rough planning, it is possible to achieve sufficient accuracy and short time. Efficient planning.

Here, as an example, for the first macro mathematical model solution, the first one day of the solution is the macro plan confirmation period, and for the second and subsequent macro mathematical model solutions, The first three days will be the macro plan final period. For the production / distribution plan obtained as a result of the simulation based on the solution to the micro mathematical model, the first 8 hours are determined throughout the planning period.
(6) All selectable processes (conveying facilities) of each product (extraction storage tank) are extracted (extraction unit 306 in FIG. 3 and step S406 in FIG. 4). In this process, as shown in FIG. 6 (a), the conveyance path is searched for the raw material storage tank, and all selectable conveyance paths for each storage tank are guided.

The details of the extraction operation of all selectable transport paths for each storage tank will be described below.
First, for the transport route search that describes the distribution structure, the yard / raw material arrangement, the material storage tank stacking brand, the reclaimer that can be used in the yard, the belt conveyor series that can be used in the reclaimer, and the belt conveyor series that can enter the raw material storage tank The information table 61 is fetched from the condition setting and fetching unit 20 of FIG.

  For example, taking the case of the raw material storage tank 2 as an example, with reference to FIG. 6B, the raw material storage tank 2 is searched from the starting facility of the transport path search information table 61 in step S61 (step 1).

  Next, in step S62 (step 2), the brand corresponding to the brand B loaded in the raw material storage tank 2 is searched from the raw material mountain brand in the conveyance path search information table 61.

  Next, in step S63 (step 3), a set of yards and reclaimers corresponding to the retrieved raw material mountain brand is retrieved. Here, it can be seen that (yard 1, RR No. 1) and (yard 2, RR No. 2) can be used.

  Next, in step S64 (step 4), a usable belt conveyor series is searched from the place where the searched starting equipment column and the searched raw material mountain brand intersect. In this case, it is understood that the series 2 can be used when (yard 1, RR No. 1) is used, and the series 4 can be used when (yard 2, RR No. 2) is used.

  As described above, the two transport paths (yard 1, RR No. 1, series 2) and (yard 2, RR No. 2, series 4) are extracted as the transport paths to the raw material storage tank 2.

  (7) Build all processes and transport paths (building unit 307 in FIG. 3, step S406 in FIG. 4). When the extraction of the conveyance path is completed for all the raw material storage tanks, the process proceeds to step S65 (step 5), and the allocation pattern of the conveyance paths is constructed with respect to the usable processes and the conveyance paths guided to the all raw material storage tanks.

In this example, for the raw material storage tank 1 (first yard, RR No.1, first belt conveyor series), for the raw material storage tank 2 (first yard, RR No.1, second Belt conveyor series) and (second yard, RR No.2, fourth belt conveyor series), for raw material storage tank 3 (first yard, RR No.1, third belt conveyor series) ) For the raw material storage tank 4, a transport path of (first yard, RR No.1, fifth belt conveyor series) and (second yard, RR No.2, sixth belt conveyor series) An allocation pattern is constructed.
(8) Formulate the macro mathematical model (macro mathematical model construction unit 308 in FIG. 3, step S407 in FIG. 4).

  In this process, a formula model is formulated with a plan creation macro time accuracy in which a plan creation macro instruction calculation period set for the constructed allocation pattern is set based on setting conditions, logistics constraints, and logistics status.

  Inter-process constraint model that describes the constraints between the processes of a series of operations of reclaimer work, transfer work, and tank entry work that occur from the start of tank entry work to the end of tank entry work, and interference within the same process Is constructed from the in-process interference constraint model.

  Define variables to determine whether or not to use reclaimers, belt conveyors, and raw material reservoirs to build an inter-process constraint model. Specifically, three variables shown in the following (formula 1) are defined.

  Where t is the elapsed time from the planning start date and time, the macro time accuracy is x hours, and the macro instruction calculation period is y hours, t takes the values of 0, x, 2x ..., yx, and each raw material Prepare y / x α variables for storage bin and reclaimer RR. Similarly, a β variable is prepared for each raw material storage bin and belt conveyor BC, and a γ variable is prepared for each raw material storage tank.

  In this embodiment, since the macro mathematical model is aimed at planning from a long-term perspective, when the macro mathematical model is created for the first time, the macro time accuracy is 1 hour, and the macro instruction calculation period is 3 days (72 Time), the range that can be taken as time t is as shown in (Equation 2) below.

  In addition, when creating a macro mathematical model for the second and subsequent times, the plan creation time accuracy is 2 hours, and the macro instruction calculation period is 3 days (72 hours), so the possible range of time t is as follows (Equation 3) become that way.

  The following description will be given by taking the case of creating a macro mathematical model for the first time as an example. In this case, the variables shown in the following (formula 4) are finally prepared from all the processes and the conveyance path construction.

  In the inter-process constraint model, a series of operations of reclaimer, transfer, and tank entry need to be performed at the same time. The constraint in this case is expressed by the following (Formula 5).

  In addition, entering the same tank should not be performed at the same time using different reclaimers and different belt conveyor systems. The constraint in this case is expressed by the following (formula 6).

  Next, the following variables and constants are defined in order to model changes in inventory.

  The change formula of the stock quantity is expressed as (Equation 8) below.

  In addition, the inventory upper and lower limit constraint is expressed by the following (formula 9).

  In the case where there is a restriction of the above (formula 9), there is a concern about occurrence of a case where there is no solution when the mathematical model is solved. When considering practical use, it is essential to prevent the absence of a solution. For this purpose, the following variables are introduced.

  In order to prevent the absence of the solution according to (Equation 9), (Equation 9) is changed to the following (Equation 11).

  In the in-process interference constraint model, (raw material storage tank 1, first yard, reclaimer RR No.1, first belt conveyor series), (raw material storage tank 3, first yard, reclaimer RR No.1, third When using the conveyor belt), it is necessary to use both RR No.1 when entering the raw material storage tank 1 and when entering the raw material storage tank 3, but this equipment Then, it can not be used when time overlaps (temporal interference). The constraint in this case is expressed as (Equation 12) below.

  Further, by transforming these equations, the equation model can be constructed as a simple linear form (Equation 13) to (Equation 15) below and an integer constraint equation.

  X is a matrix representation of the operation judgment of each facility, raw material storage tank inventory, etc., A and B are predetermined determinants, Xmin and Xmax are matrix representations of the lower and upper limit levels of the defined variables, respectively ( Equation 15) indicates that all or some of the elements of x are integer constraints.

  Here, a case where the planner intentionally designates a part of production / distribution in advance will be described. For example, when it is desired to specify in advance to enter the raw material storage tank 1 at time 2x and into the raw material storage tank 2 at time 10x using the reclaimer No2, the following (Equation 16) is added as a constraint equation.

  (9) The macro mathematical model is optimized based on the macro evaluation function (the solution finding unit 309 in FIG. 3 and step S408 in FIG. 4). In this process, LP (linear programming), MIP (mixed integer programming), QP (based on a macro evaluation function set in advance) for each of the macro mathematical model models composed of the linear and integer constraint equations constructed as described above. By solving the problem as an optimization problem using a mathematical programming method such as quadratic programming) or a method combining mathematical programming with a heuristic method such as tab search, genetic algorithm, etc. The process route used for the tank operation and the production / logistics plan for the approximate tank entry time are required.

  For example, in the above optimization calculation, if a level that forms a sub-optimal solution is acceptable, a genetic algorithm is used to form each integer variable I as a gene, and I formed by the genetic algorithm is a determined value. After that, can be solved as an LP problem. When it is desired to obtain an optimal solution, it is solved as a mixed integer programming problem.

  Here, an example in the case of using the line format for the macro evaluation function is shown. In this embodiment, the purpose is to achieve a high level of stock level stability and efficient use of equipment. Therefore, the macro evaluation function is closer to the target stock amount specified by the operator, and the lower the equipment utilization rate, the better. Is a function to obtain The target stock amount and the short stock amount from the target stock amount are defined as shown below.

  The relationship between the target stock amount and the shortage amount from the target stock amount is expressed by the following (formula 18).

  When the evaluation function is expressed by an equation, the following (Equation 19) is obtained.

  When considering practical use, it is necessary to avoid overflowing the inventory from the upper limit of inventory and interruption from the lower limit of inventory as much as possible. Therefore, (Equation 19) representing the evaluation function is changed to the following (Equation 20).

  An optimal solution is obtained by solving the above formulated formula (formula model) by the mixed integer programming.

  Here, a solution result for the macro mathematical model will be described with reference to a diagram (FIG. 7) illustrating an example of a solution result for the macro mathematical model in one embodiment of the present invention.

  FIG. 7 shows, as an example, the solution result for the macro mathematical model created for the first time, and shows the tank destination, process route, and tank entry time for three days with an accuracy of one hour unit. For example, for the raw material storage tank 1, use the (first yard, RR No.1, first belt conveyor series) to enter the tank between 0 and 2 hours of the macro instruction calculation period. For storage tank 2, use (second yard, RR No.2, fourth belt conveyor system) to enter tank between 0 and 5 hours of macro instruction calculation period, On the other hand, using (first yard, RR No.1, third belt conveyor system), the tank is entered during 2 to 9 hours of the macro instruction calculation period. (Second yard, RR No. 2, sixth belt conveyor series) is used to enter the tank for 24 to 27 hours in the macro instruction calculation period.

  (10) Based on the solution obtained by the macro mathematical model, the micro mathematical model is formulated (micro mathematical model building unit 310 in FIG. 3, step 409 in FIG. 4). In this process, based on the solution obtained by the macro mathematical model, the mathematical model is formulated with the micro time accuracy set for the micro instruction calculation period set based on the setting conditions, physical distribution constraints, and physical distribution status.

  In one embodiment of the present invention, a micro mathematical model with minute accuracy required in actual operation is constructed based on a solution obtained with a macro accuracy of one hour. The micro mathematical model is an inter-process constraint model that describes the inter-process constraints of the reclaimer work, transfer work, and tank entry work that occur from the start of the tank entry work to the end of each tank. It is constructed from an in-process interference constraint model that models interference in the process.

  From the results of solving the macro mathematical model, a series of reclaimer work, transfer work, and tank entry work that occurs within the microinstruction calculation period and the process route for performing the work are determined. Define variables for the work and process path. The operation start time and end time of the reclaimer are defined as shown in (Equation 21).

  The transport start time and end time of the belt conveyor series are defined as shown in (Equation 22).

  The entry time and end time are defined as shown in (Equation 23).

  For example, for the raw material storage tank 1, use the (first yard, RR No.1, first belt conveyor series) to enter the tank between 0 and 2 hours of the macro instruction calculation period. For storage tank 2, use (second yard, RR No.2, fourth belt conveyor system) to enter tank between 0 and 5 hours of macro instruction calculation period, On the other hand, using (first yard, RR No.1, third belt conveyor system), the tank is entered during 2 to 9 hours of the macro instruction calculation period. (4th RR No.2, 6th belt conveyor series) using 4 series of work to enter the tank during 24 to 27 hours of macro instruction calculation period, The following variables are defined as work 1, work 2, work 3, and work 4.

  The inter-process constraint model is shown below. Reflecting the rough time of 1-hour unit accuracy obtained from the solution result for the macro mathematical model, the constraints on the range that can be taken by the accurate work start time and end time are formulated. For example, it is assumed that a deviation from the rough time of 1 hour unit accuracy obtained from the solution result for the macro mathematical model is allowed for w hours at both the start and end.

  Here, it is assumed that the start time and end time of the reclaimer for a series of operations obtained from the solution to the macro mathematical model are given by constants shown in (Equation 25) below.

  The constraint in this case is expressed by the following (Equation 26).

  There is a certain time lag (l, m, n, and p are constants) between the processes for a series of operations including reclaimer work, transport work, and tank entry work. The constraint in this case is expressed by the following (formula 27).

  In addition, the tank stock level at the start of entry of the raw material storage tank and the tank stock level at the end of entry are defined as shown in (Equation 28).

  When the amount of tanks in and out of the raw material storage tank is constant regardless of time, the relational expression between the start and end times of the reclaimer and the tank stock level of the raw material storage tank is expressed as (Equation 29) below. Is done.

  Also, the tank entry time must be earlier than the tank entry time. The constraint in this case is expressed as (Equation 30) below.

Furthermore, Iriso starting tank inventory levels ^{_{R bin (t s JOB, RR}} ) is the lowest level generally in the convenience of operation management R _{bin, sL} (management lower limit value) or more, Iriso end of the tank inventory levels R _bin (t ^e _{JOB, RR)} highest R _bin, or less is required _eU (upper control limit) in the. The constraint in this case is expressed as in (Equation 31) below.

  In the in-process interference constraint model, it is necessary to provide constraints so that work times that may use the same equipment at the same time (temporal interference) do not overlap. For example, in case of work 1 (raw material storage tank 1, yard 1, RR No.1, series 1) and work 3 (raw material storage tank 3, yard 1, RR No.1, series 3), both RR No. 1 are used. In addition, there is a possibility that the working time may overlap, but the same equipment cannot be used when the time overlaps. The constraint in this case is expressed as in (Equation 32) below.

  Further, by transforming these equations, the equation model can be constructed as a simple linear form of the following (Equation 33) to (Equation 34).

X is a matrix representation of the operation start / end time of each facility, raw material storage tank inventory, etc., A and B are predetermined determinants, and Xmin and Xmax are matrix representations of the lower and upper limit levels of the defined variables, respectively. Is.
(11) The micro mathematical model is optimized based on the micro evaluation function (the micro solicitation unit 311 in FIG. 3 and step S410 in FIG. 4). In this process, LP (linear programming), MIP (mixed integer programming), QP (based on a preset micro-evaluation function for each of the above-constructed micro mathematical model expressions composed of linear and integer constraint expressions. Optimized entry and exit times by solving the problem as an optimization problem using a combination of mathematical programming such as secondary programming) or meta-heuristic techniques such as tab search and genetic algorithms and mathematical programming , The tank amount, the reclaimer operation start time, the reclaimer operation end time, the transport belt conveyor system operation start time, and the transport belt conveyor system operation end time are calculated.

  The payout raw material pile, payout yard, reclaimer to be used, and belt conveyor series to be used are already determined from the solution result for the macro mathematical model. The process route to be used and the tank entry time will be determined.

Here, an example in the case of using a linear format for the micro evaluation function is shown. In this embodiment, the purpose is to stabilize the stock level at a high level, so that the target evaluation function enters the tank entry target level R _{bin, sr} and the tank entry target level R _{bin, er} specified by the operator. bath starting level ^{_{R bin (t s JOB, RR}} ), Iriso End level ^{_{R bin (t e JOB, RR}} ) is a function to obtain a good value closer. When the evaluation function is expressed by an equation, the following (Equation 35) is obtained.

  In order to linearize the evaluation function of the above equation, dividing the tank start level and the tank end level into two variables before and after each target level as in the following formula, the following (formula 36) and Become.

  When the above equation is used, the evaluation function has a linear form as shown in (Equation 37) below.

  By solving the above formulated formula (formula model) by linear programming, an optimal solution can be obtained for each formula model.

  (12) A simulation is performed based on the obtained solution (simulation unit 312 in FIG. 3, step S411 in FIG. 4). Solution to the mathematical model obtained by the above micro mathematical model and raw material acceptance plan, raw material yard plan, equipment repair plan, raw material yard present state, tank stock present state, tank cut-out amount present state, equipment operation・ Set all or part of the target manufacturing process / conveyance process route / conveyance route / product / equipment / moving body based on the failure status and all or part of the operating preconditions from the operator Simulation is performed with the accuracy of the plan creation simulation in which the micro plan finalization period is set.

  In this simulation, constraints and operating rules that could not be incorporated into the macro mathematical model and the micro mathematical model are also incorporated into the simulation, and the simulation results are used to calculate the solution obtained as the solution to the micro mathematical model. Change to a production and logistics plan that can be used without problems.

  For example, in order to realize this simulation, a simulator capable of simulating the manufacturing process, the transport process route, the transport route, the product, the equipment, and the operation / status of the moving body in detail may be used.

  This makes it possible to create a production / logistics plan that takes into account the time accuracy required for actual operation and the fine constraints required for actual operation.

  In addition, examples of constraints that are difficult to handle with mathematical models include the setup time as a function of distance, the reclaimer operation, and the entry to the raw material storage tank when moving the reclaimer when the raw material storage tank changes. Incorporating a time lag of several minutes or so into the simulation and accurately simulating it makes it possible to create a production / logistics plan that takes into account the fine constraints required for actual operation.

  (13) The production / distribution plan is confirmed (confirmation unit 313 in FIG. 3, step S412 in FIG. 4). In this process, the micro plan determination period set in the production / distribution plan derived by the simulation is determined.

  As shown in the drawing (FIG. 8) for explaining the production / distribution plan creation procedure in this embodiment, the first eight hours of the created production / distribution plan are determined. Of the created production / distribution plan, the part that does not enter the micro plan finalization period is discarded without being finalized.

  (14) It is determined whether plans for the plan creation period or the macro plan determination period have been determined (the determination unit 314 in FIG. 3 and steps S413 and 414 in FIG. 4). In this process, it is determined whether the plan decision period established up to that point in time has confirmed a preset macro plan decision period or created a plan creation period.

  In the present embodiment, the plan for the macro plan determination period is determined when the plan is determined in the third micro loop.

  Further, when the plan is confirmed in the micro 12th loop, the plan for the plan creation period is confirmed. For this reason, a production / distribution plan for four days is created when the plan is finalized in the twelfth loop, and the process is terminated.

  (15) The planning start date is updated (update unit 315 in FIG. 3, steps S415 and 416 in FIG. 4). If the determined plan determination period has not been determined for the preset plan creation period, the date immediately after the production / distribution plan period determined in the production / distribution plan is set as a new planning start date.

  In this embodiment, as shown in FIG. 8, the planning start date that was initially 0 o'clock in the first loop in the first loop is 8 o'clock on the first day, and the planning start that was initially 8 o'clock in the second loop is started. Update the day to 16:00 on the first day.

  Then, if the plan finalization period that has been determined up to that point is the predetermined macroplan finalization period, the process of step S406 in FIG. 4 is executed, otherwise the process of step S409 in FIG. 4 is executed.

  In the present embodiment, when the planning start date and time is updated after the completion of the micro third loop, the process proceeds to step S406 in FIG. 4, and otherwise, the process proceeds to step S409 in FIG.

  As described above, according to the present embodiment, the brands A, B, and C to be transported to the raw material storage tanks 1 to 4 are detected according to the current distribution state, and the long-term detection is performed on the detected brands A, B, and C. From a macro view, build a macro formula model with a predetermined plan creation time accuracy, solve the built macro formula model based on the macro evaluation function, and take a detailed micro view from the solved solution, Build a micro mathematical model with a predetermined plan creation time accuracy, solve the constructed micro mathematical model based on the micro evaluation function, and simulate the operation and status of the reclaimer, belt conveyor series, and raw material storage tank based on the solved solution In the production / logistics plan obtained from the simulated results, the date immediately after the set plan confirmation period is set as the new planning start date, so that the Since the series of processes are sequentially performed so that determining the plan, production requiring any time accuracy, logistics planning, can be fast and optimize in detail, and can be applied to actual operation as such.

  In addition, although embodiment mentioned above demonstrated the case where this invention was applied to a tank introduction plan preparation apparatus, it is also possible to apply to a raw material physical distribution control apparatus. In this case, an instruction is given to the control device of the actual plant based on the created tank entry plan.

  In this way, the actual plant has an optimal order of tank entry, tank start and end times, tank volume, reclaimer operation start time, reclaimer operation end time as well as discharged raw material mountain, yard, used reclaimer, transfer The raw material yard operation is executed according to the belt conveyor system and the incoming tank raw material storage tank.

  When the actual plant is controlled based on the tank entry plan created in this way, the current physical distribution state in the actual plant changes, so that information is taken out at certain time intervals and supplied to the condition setting and taking-in unit 20. .

  The tank entry plan creation unit 21 takes in input data, sets initial values, and sets conditions using input data, initial values, and condition setting functions (step S401 in FIG. 4).

  Then, a macro mathematical model is constructed based on the given current distribution state and distribution restrictions.

  The macro mathematical model is solved based on the macro evaluation function. A micro mathematical model is constructed based on the result of the solution. The micro mathematical model is solved based on a micro evaluation function. Based on the results of the solution, the operation and situation of the reclaimer, belt conveyor system and raw material storage tank are simulated. Of the production / logistics plan obtained from the simulated results, the micro plan finalization period is confirmed, the planning start date is updated to the date and time immediately after the micro plan finalization period, and the plan finalization period is the macro plan finalization period. If it is confirmed, return to the formula formulation of the macro formula model, if it is not settled for the macro period, return to the formula processing of the micro formula model, and repeat until the end of the planning period, the order of tank entry, the start and end of the tank Not only the time, tank amount, reclaimer operation start time, and reclaimer operation end time, but also the payout raw material mountain, yard, used reclaimer, conveyor belt conveyor series, and incoming tank raw material storage tank are determined. By controlling the raw material distribution based on this result, the control of the raw material distribution is optimally executed.

-Second Embodiment-
Further, as a second embodiment, a macro mathematical model for processing details of the tank entry plan creation unit 21 and an embodiment that does not use the micro mathematical model are the details of the production / distribution plan creation device (the tank entry plan creation unit) 21. Fig. 10 shows a simple configuration (Fig. 10), a flowchart (Fig. 11) showing detailed processing contents of the production / distribution plan creation device (incoming tank plan creation unit) 21, and production in the case of using the example of Fig. 5 -It demonstrates in detail using the figure (FIGS. 12-14) which shows the detail of operation | movement inside the physical distribution plan preparation apparatus (entrance planning part) 21. FIG.

Here, the simulation process is step S1111 in FIG. 11, the macro gene model construction process is step S1107 in FIG. 11, the macro gene model optimization process is step S408 in FIG. 11, the micro mathematical model construction process is step S1109 in FIG. The priority model construction process includes step S1110 in FIG. 11, the production / distribution plan determination process includes step S1112 in FIG. 11, the micro update process includes step S1115 in FIG. 11, and the macro execution determination process includes steps S1114 and S1116 in FIG. The
(1) Capture input data, set initial values and conditions (data capture unit 1001 in FIG. 10, step S1101 in FIG. 11). Processing similar to that of the data fetch unit 301 in FIG. 3 and step S401 in FIG. 4 is performed.

  (2) A production / distribution plan creation period is set (plan creation period setting unit 1002 in FIG. 10, step S1102 in FIG. 11). Processing similar to the plan creation period setting unit 302 in FIG. 3 and step S402 in FIG. 4 is performed.

  (3) The production / distribution plan creation time accuracy is set (plan creation time accuracy setting unit 1003 in FIG. 10, step S1103 in FIG. 11). In this process, macro time accuracy, micro time accuracy and simulation accuracy for creating a production / distribution plan are set. The macro time accuracy, micro time accuracy, and simulation accuracy can be set arbitrarily according to the needs of the planner.

  For example, the macro time accuracy is coarse, and after roughly determining the production / logistics plan from a long-term perspective, the production / logistics plan is determined in detail with finer micro-time accuracy based on the decision result. Therefore, it is possible to create an efficient plan with sufficient accuracy and a short time from a long-term perspective.

  As another example, sufficient accuracy and short time can be achieved by reducing the accuracy in the first half of the planning period that requires fine planning accuracy, and coarsening the accuracy in the second half of the sufficient planning period. Can create efficient plans.

  Here, as an example, when creating a macro gene model for the first time, the macro time accuracy is set to 1 hour, and when creating the second and subsequent macro gene models, the macro time accuracy is set to 2 hours. . In the micro time accuracy and simulation part, the accuracy is made minute throughout the planning period. This makes it possible to create a production / logistics plan with the accuracy that can be applied as it is in actual operation by building a model that can be solved accurately in a short time and performing simulation with the accuracy required in actual operation. .

  (4) A production / distribution instruction calculation period is set (instruction calculation period setting unit 1004 in FIG. 10, step S1104 in FIG. 11).

  Processing similar to the instruction calculation period setting unit 304 in FIG. 3 and step S404 in FIG. 4 is performed.

  (5) A production / distribution plan decision period is set (plan decision period setting unit 1005 in FIG. 10, step S1105 in FIG. 11).

  In this process, a macro plan finalizing period and a micro plan finalizing period for finalizing the production / distribution plan are set. The macro plan finalization period and the micro plan finalization period can be arbitrarily set as required by the planner.

  For example, by shortening the plan decision period in the first half of the plan creation period that requires fine planning accuracy, and increasing the plan decision period in the second half of the plan creation period that is sufficient for rough planning, it is possible to achieve sufficient accuracy and short time. Efficient planning.

  Here, as an example, for the first macrogene model solution, the first one day of the solution is the macro plan finalization period, and for the second and subsequent macrogene model solutions, The first three days will be the macro plan final period. For the production and logistics plan obtained as a result of simulation based on the solution for the micro priority model, the first 8 hours are determined throughout the planning period.

  (6) All selectable processes (conveying facilities) of each product (extraction storage tank) are extracted (extraction unit 1006 in FIG. 10, step S1106 in FIG. 11). The same processing as the extraction unit 306 in FIG. 3 and step S406 in FIG. 4 is performed.

  (7) Build all processes and transport paths (building unit 307 in FIG. 3, step S406 in FIG. 4). Processing similar to that of the construction unit 307 in FIG. 3 and step S406 in FIG. 4 is performed.

  (8) The macro gene model is constructed (macro gene model construction unit 1008 in FIG. 10, step S1107 in FIG. 11). In this process, the gene model is constructed with the plan creation macro time accuracy in which the plan creation macro instruction calculation period set for the constructed allocation pattern is set based on the setting conditions, the distribution constraints, and the distribution status. In order to construct a gene model, a variable that determines whether or not to use a reclaimer, a belt conveyor, and a tank is defined as a gene. Specifically, the variable shown in the following (formula 38) is defined as a gene.

  Where t is the elapsed time from the planning start date and time, the macro time accuracy is x hours, and the macro instruction calculation period is y hours, t takes the values of 0, x, 2x ..., yx, and each raw material Prepare gene γ for storage bin.

  In this embodiment, since the macro gene model is aimed at planning from a long-term perspective, when the macro mathematical model is created for the first time, the macro time accuracy is 1 hour, and the macro instruction calculation period is 3 days (72 Time), the range that can be taken as time t is as shown in (Equation 39) below.

  Further, in the case of creating the macro mathematical model after the second time, the plan creation time accuracy is 2 hours, and the macro instruction calculation period is 3 days (72 hours). Therefore, the range that can be taken as the time t is as follows (Formula 40) become that way.

The following explanation will be given by taking the case of creating a macro gene model for the first time as an example.
In this case, the gene shown in the following (formula 41) is finally prepared from all the steps and the construction of the transport route.

  Obtaining a macro gene model is equivalent to determining the value of each gene in the above (formula 41). For this reason, as shown in FIG. 12, a group of the genes of the above (formula 41) is defined as a chromosome and one having one chromosome as one individual. Here, one individual corresponds to a solution of one macro gene model.

  (9) The solution is obtained based on the macro gene model (the solution obtaining unit 1009 in FIG. 10, step S1108 in FIG. 11).

  In this process, a sub-optimal solution is obtained using a genetic algorithm based on the constructed macro gene. Here, the details of the solution processing in the solution calculation unit will be described based on a diagram (FIG. 13) showing a flowchart of detailed processing.

  (9-1) An initial population is generated (step S1301 in FIG. 13). Create an initial population. Here, a random number is used as an example. That is, a random number is generated for each gene of each individual, and if it is less than 0.5, 0 is set for each gene, and if it is 0.5 or more, 1 is set for each gene. However, entering the same tank should not be performed at the same time using different reclaimers and different belt conveyor systems.

  For this reason, when the gene at a certain time in a certain tank is 1, the above-mentioned tank, the same tank as the time, another reclaimer at the same time, and another belt conveyor gene are not operated with random numbers. Set. By repeating this operation, about 1000 individuals (individual groups) are generated.

  (9-2) The fitness is evaluated (step S1302 in FIG. 13). The fitness of all individuals (populations) is evaluated. Here, it is evaluated whether the following conditions are satisfied.

<Macro constraint 1>
When the gene at a certain time in a certain tank is 1, the genes of the above tank, the same time and the same tank, another reclaimer at the same time, and another belt conveyor are 0.

<Macro constraint 2>
When a certain reclaimer and a gene at a certain time are set to 1, the genes of the above reclaimer, the same reclaimer as the time, the separate tank and the different belt conveyor at the same time are zero.

<Macro restriction 3>
When the gene of a certain belt conveyor and a certain time is 1, the genes of the above belt conveyor, the same time and the same belt conveyor, the same tank, the different reclaimer are 0.

  Next, the stock is evaluated. First, the following variables and constants are defined.

  The change equation of the stock quantity is expressed as (Equation 43) below.

  In addition, the following variables are introduced to take into account the difference between the stock quantity and the stock quantity.

  The relationship between (Equation 43) and (Equation 44) is (Equation 45) described below.

The evaluation relating to inventory is the following evaluation 1.
Macro evaluation 1: The amount of S overflow ^t _bin and S supplement ^t _bin is used as an evaluation value. In the present embodiment, since the purpose is to use the efficient equipment, it is necessary to obtain a better evaluation value as the operating rate of the equipment is lower.

For this reason, the following evaluation 2 is introduced.
Macro evaluation 2: The number of genes for which a value of 1 is set for a gene is used as an evaluation value.

  (9-3) A selection is made based on the fitness evaluation (step S1303 in FIG. 13). About 100 individuals are selected in order of increasing total of the evaluation value of the macro evaluation 1 and the evaluation value of the macro evaluation 2 satisfying the introduced macro conditions 1 to 3 of the fitness.

  (9-4) A crossover operation is performed on the selected individual group (step S1304 in FIG. 13). Select any two individuals from the selected group of individuals and perform crossover to create a new individual. In the crossover of the present embodiment, the first half of the chromosome of one individual and the latter half of the chromosome of the other individual are combined to create a new individual. This operation is repeated until the number of individuals reaches about 1000.

  (9-5) Mutation operation is performed on the individual group (step S1305 in FIG. 13). Select one individual from the group of individuals, and change any gene generated by random numbers to a value different from the current value. Repeat this operation about 100 times.

  (9-6) An end determination is made (step S1306 in FIG. 13). When the number of times (9-2) to (9-5) reaches a preset number of times, or when the fitness falls below a predetermined value, the gene manipulation process is terminated. If not reached, the process returns to the process (9-2), and the processes (9-2) to (9-6) are repeated.

  Here, when it is determined that the process is finished, an individual having the smallest evaluation value in the individual group is determined as a solution.

  The solution result for the macro gene model will be described with reference to a diagram (FIG. 14) showing an example of the solution result for the macro gene model in one embodiment of the present invention.

  FIG. 14 shows, as an example, the solution result for the macro gene model created for the first time, and shows the tank destination, process route, and tank entry time for three days with an accuracy of one hour unit. For example, for the raw material storage tank 1, use the (first yard, RR No.1, first belt conveyor series) to enter the tank between 0 and 2 hours of the macro instruction calculation period. For storage tank 2, use (second yard, RR No.2, fourth belt conveyor system) to enter tank between 0 and 5 hours of macro instruction calculation period, On the other hand, using (first yard, RR No.1, third belt conveyor system), the tank is entered during 2 to 9 hours of the macro instruction calculation period. (Second yard, RR No. 2, sixth belt conveyor series) is used to enter the tank for 24 to 27 hours in the macro instruction calculation period.

  (10) Based on the solution obtained by the macro gene model, a micro priority model is constructed (micro priority model construction unit 1010 in FIG. 10, step 1109 in FIG. 11). In this process, based on the solution obtained by the macro gene model, the micro priority model is constructed with the micro time accuracy in which the micro instruction calculation period set based on the setting condition, the distribution constraint, and the distribution situation is set.

  In one embodiment of the present invention, a micro-priority model with minute accuracy required in actual operation is constructed based on a solution obtained with a macro accuracy of one hour.

  The micro priority model defines the start and end times of a series of reclaimer work, transfer work, and tank entry work used from the start of tank entry work to the end of tank entry as variables with minute accuracy. . Specifically, from the results of the solution for the macro mathematical model, a series of work of reclaimer work, transport work, tank entry work that occurs within the micro instruction calculation period, and the process route for performing the work are determined. The process route is identified, and the variables for the operation and the process route are defined.

  The start time and end time of the work JOB that enters the tank bin using the reclaimer RR and the belt conveyor BC are defined as shown in (Equation 46).

  For example, for the raw material storage tank 1, use the (first yard, RR No.1, first belt conveyor series) to enter the tank between 0 and 2 hours of the macro instruction calculation period. For storage tank 2, use (second yard, RR No.2, fourth belt conveyor system) to enter tank between 0 and 5 hours of macro instruction calculation period, On the other hand, using (first yard, RR No.1, third belt conveyor system), the tank is entered during 2 to 9 hours of the macro instruction calculation period. (4th RR No.2, 6th belt conveyor series) using 4 series of work to enter the tank during 24 to 27 hours of macro instruction calculation period, The following variables are defined as work 1, work 2, work 3, and work 4.

  In addition, reflecting the rough time of 1-hour unit accuracy obtained from the solution result for the macro mathematical model, the range of accurate start time and end time of minute accuracy is the time obtained from the macro result. From within a certain time range. For example, it is assumed that the deviation from the rough time of 1 hour unit accuracy obtained from the solution result for the macro mathematical model is allowed for 1 hour at both the start and the end.

  Here, it is assumed that the start time and the end time for a series of operations obtained from the solution to the macro mathematical model are given by the constants shown in (Formula 48) below.

  (11) Obtain a solution based on the micro priority model (micro solution unit 1011 in FIG. 10, step S1110 in FIG. 11).

  In this process, a suboptimal solution is obtained using a local search algorithm based on the constructed micro priority model.

  Here, the details of the solution processing in the solution calculating unit will be described based on a diagram (FIG. 15) showing a flowchart of detailed processing.

  (11-1) An initial solution is generated (step S1501 in FIG. 15). When executing a local search, a solution is generated as an initial starting point.

  Here, in generating the initial solution, the initial solution needs to satisfy the following constraints that the solution should satisfy. In addition, the following constraints must be satisfied when searching for neighborhoods of solutions in a local search.

<Micro constraint 1>
For each task, the start and end time deviations with a rough start and end time of 1 hour unit accuracy obtained from the solution to the macro mathematical model are within one hour.

<Micro constraint 2>
When work is performed at a certain time in a certain tank, the work is not performed on the left tank at the same time, with another reclaimer and another belt conveyor.

<Micro constraint 3>
When work is being performed at a certain time on a belt conveyor, the work is not performed in a separate tank or separate reclaimer at the same time on the left belt conveyor.

  In the present embodiment, a macro solution is used as an initial solution. That is, it is determined as in the following (formula 49).

  (11-2) An evaluation value is calculated (step S1502 in FIG. 15). Evaluation of the initial solution is performed using the evaluation value.

  In the present embodiment, the amount of lower limit cracking and upper limit overflow with respect to the preset upper and lower limits of the inventory amount is evaluated as an evaluation value, and the smaller the value, the better the evaluation.

  Here, the tank stock level at the start of tank entry of the raw material storage tank and the tank inventory level at the end of tank entry are defined as shown in (Equation 50).

  When the amount of tanks in and out of the raw material storage tank is constant regardless of the time, the relational expression between the start and end times of the reclaimer and the tank stock level of the raw material storage tank is expressed as (Equation 51) below. Is done.

Iriso starting tank inventory levels ^{_{R bin (t s JOB, RR}} , BC, bin) is generally the lowest level R _bin with the convenience of operation _{management, sL} (management lower limit value) or more, the tank stock at Iriso End level ^{_{R bin (t e JOB, RR}} , BC, bin) is the highest level R _{bin with,} it is desired that a _eU (upper control limit). For this reason, let the total value of the following values be an evaluation value.

Micro evaluation 1: Tank stock level R _bin (t ^s _{JOB, RR, BC, bin} ) at the start of tank entry is the amount that the minimum level R _{bin, sL} is interrupted Micro evaluation 2: Tank stock level R _bin at the end of tank entry _{^{(t e JOB, RR, BC}} , bin) is the highest level R _bin, with respect to the initial solution the sum of the amount the micro evaluation 1 and micro evaluation 2 beyond the _eU, is calculated, and the evaluation value for the initial solution .

  (11-3) Search the vicinity (step S1503 in FIG. 15). Search for a neighborhood that satisfies the introduced micro conditions 1 to 3 and where the sum of the evaluation value of micro evaluation 1 and the evaluation value of micro evaluation 2 is small (another solution in which the solution is changed by a small amount).

  In the present embodiment, one or both of the start time and the end time of the work time is reduced by 1 in the direction of reducing the amount of interrupting the lowest level (advancing the start time) or reducing the amount exceeding the maximum level (delaying the end time). Change minutes. However, even in this change, the micro conditions 1 to 3 need to be satisfied.

  (11-4) Update the solution (step S1504 in FIG. 15). In the neighborhood search, when a solution with an improved evaluation value is found, this solution is registered as a new solution.

  (11-5) An end determination is made (step S1505 in FIG. 15). When the processes of (11-3) to (11-4) reach a preset number of times or when a neighborhood where the evaluation value is improved is not found, the neighborhood search process is terminated. If the above condition is not reached, the processes (11-3) to (11-4) are repeated.

  Here, when it is determined that the process is finished, the last updated solution having the smallest evaluation value is determined as the solution.

  (12) A simulation is performed based on the obtained solution (simulation unit 1012 in FIG. 10, step S1111 in FIG. 11).

  Solution to the above micro priority model, raw material acceptance plan, raw material yard plan, equipment repair plan, raw material yard status, tank inventory status, tank extraction status status, equipment operation / failure status, Based on all or part of the operational preconditions from the operator, the micro plan confirmation period in which all or part of the target manufacturing process / conveyance process route / conveyance route / product / equipment / moving body is set Simulate with planning simulation accuracy with minutes set.

  In this simulation, constraints that could not be incorporated into the macro gene model and the micro priority model, operation rules, etc. were also incorporated into the simulation, and by simulation, the solution obtained as the solution to the micro priority model was actually operated. Change to a production and logistics plan that can be used without problems.

  For example, in order to realize this simulation, a simulator capable of simulating the manufacturing process, the transport process route, the transport route, the product, the equipment, and the operation / status of the moving body in detail may be used.

  This makes it possible to create a production / logistics plan that takes into account the time accuracy required for actual operation and the fine constraints required for actual operation.

  In addition, examples of constraints that are difficult to handle with the genetic model and priority model include the setup time as a function of distance, the reclaimer operation from the reclaimer operation to the raw material storage tank when the raw material storage tank subject to entry changes. By taking a time lag of several minutes before entering the tank into the simulation and accurately simulating it, it becomes possible to make a production / logistics plan that takes into account the fine constraints required for actual operation.

  (13) The production / distribution plan is confirmed (confirmation unit 1013 in FIG. 10, step S1112 in FIG. 11). In this process, the micro plan determination period set in the production / distribution plan derived by the simulation is determined.

  (14) It is determined whether plans for the plan creation period or the macro plan determination period have been determined (determination unit 1014 in FIG. 10 and steps S1113 and S1114 in FIG. 11). In this process, it is determined whether the plan decision period established up to that point in time has confirmed a preset macro plan decision period or created a plan creation period.

  (15) The planning start date is updated (update unit 1015 in FIG. 10, steps S1115 and S1116 in FIG. 11). If the determined plan determination period has not been determined for the preset plan creation period, the date immediately after the production / distribution plan period determined in the production / distribution plan is set as a new planning start date.

  Then, if the plan finalization period that has been determined up to that point is the predetermined macroplan finalization period, the process of step S1106 in FIG. 11 is executed, otherwise the process of step S1109 in FIG. 11 is executed.

  As described above, according to the present embodiment, the brands A, B, and C to be transported to the raw material storage tanks 1 to 4 are detected according to the current distribution state, and the long-term detection is performed on the detected brands A, B, and C. A macro gene model is constructed with a predetermined macro timeline and a predetermined plan creation time, and a solution is obtained based on the constructed macro gene model. Based on the obtained solution, a detailed micro field of view is created and a predetermined plan is created. Based on the micro-micro priority model with time accuracy, the operation and status of the reclaimer, belt conveyor series, and raw material storage tank are simulated based on the solved solution, and the production / logistics plan obtained from the simulated results In the above, by setting the date and time immediately after the set plan confirmation period as the new planning start date and time, a series of processes for finalizing the production / logistics plan for the new target period can be performed sequentially. Production which require time accuracy will, logistics planning, can be optimized fast and detailed, and can be applied to actual operation as such.

  In addition, although embodiment mentioned above demonstrated the case where this invention was applied to a tank introduction plan preparation apparatus, it is also possible to apply to a raw material physical distribution control apparatus. In this case, an instruction is given to the control device of the actual plant based on the created tank entry plan.

  In this way, the actual plant has an optimal order of tank entry, tank start and end times, tank volume, reclaimer operation start time, reclaimer operation end time as well as discharged raw material mountain, yard, used reclaimer, transfer The raw material yard operation is executed according to the belt conveyor system and the incoming tank raw material storage tank.

  When the actual plant is controlled based on the tank entry plan created in this way, the current physical distribution state in the actual plant changes, so that information is taken out at certain time intervals and supplied to the condition setting and taking-in unit 20. .

  The tank entry plan creation unit 21 takes in input data, sets initial values, and sets conditions using input data, initial values, and condition setting functions (step S401 in FIG. 4).

  Then, a macro gene model is constructed based on the current distribution conditions and distribution restrictions.

  Solved based on macrogene model. Build a micro priority model based on the results of the solution. Solved based on micro priority model. Based on the results of the solution, the operation and situation of the reclaimer, belt conveyor system and raw material storage tank are simulated. Of the production / logistics plan obtained from the simulated results, the micro plan finalization period is confirmed, the planning start date is updated to the date and time immediately after the micro plan finalization period, and the plan finalization period is the macro plan finalization period. If it is confirmed, it will return to the construction process of the macro priority model if it is not confirmed for the macro period, and it will return to the construction process of the micro priority model and repeat until the end of the planning period, The amount of incoming tank, the start time of reclaimer operation, and the end time of reclaimer operation are determined, as well as the discharge raw material mountain, the yard, the used reclaimer, the conveyor belt conveyor series, and the incoming tank raw material storage tank. By controlling the raw material distribution based on this result, the control of the raw material distribution is optimally executed.

  In addition, the above-mentioned basin plan preparation part 21 is comprised by the microcomputer which consists of CPU (central processing unit), RAM (random access memory), ROM (read-only memory) etc., for example, a personal computer etc. It can be realized by a computer.

(Other embodiments of the present invention)
In order to operate various devices to realize the functions of the above-described embodiments, program codes of software for realizing the functions of the above-described embodiments are provided to an apparatus or a computer in the system connected to the various devices. What is implemented by operating the various devices according to a program supplied and stored in a computer (CPU or MPU) of the system or apparatus is also included in the scope of the present invention.

  In this case, the program code of the software itself realizes the functions of the above-described embodiments, and the program code itself and means for supplying the program code to the computer, for example, the program code are stored. The recorded medium constitutes the present invention. As a recording medium for storing the program code, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

  Further, by executing the program code supplied by the computer, not only the functions of the above-described embodiments are realized, but also the OS (Operating System) or other application software in which the program code is running on the computer, etc. It goes without saying that the program code is also included in the embodiment of the present invention even when the functions of the above-described embodiment are realized in cooperation with the embodiment.

  Further, after the supplied program code is stored in the memory provided in the function expansion board of the computer or the function expansion unit connected to the computer, the CPU provided in the function expansion board or function expansion unit based on the instruction of the program code Needless to say, the present invention includes the case where the functions of the above-described embodiment are realized by performing part or all of the actual processing.

It is a schematic diagram of the raw material yard manufacturing process (conveyance) which is one implementation object of the present invention. It is a figure which shows the relationship of the structure of the raw material yard entrance tank plan preparation measure using the production and distribution plan preparation apparatus by one Embodiment of this invention, and another system. It is a figure which shows the detailed structure of the production and distribution plan preparation apparatus by one Embodiment of this invention. It is a flowchart which shows the detailed processing content of the production and distribution plan preparation apparatus by one Embodiment of this invention. It is the figure which showed the simple example which reduced the scale of the raw material yard manufacturing process (conveyance) used in order to demonstrate the process outline | summary of the tank plan preparation part by one Embodiment of this invention. It is the figure for demonstrating the search of the process path | route and conveyance path | route in one Embodiment of this invention, and a flowchart which shows the search method. It is a figure which shows the example of the solution result with respect to the macro mathematical model in one Embodiment of this invention. It is a figure explaining the production and plan preparation procedure in one Embodiment of this invention. It is a block diagram explaining an example of a structure of the production and distribution plan creation apparatus by one Embodiment of this invention. It is a figure which shows the detailed structure of the production and distribution plan preparation apparatus by another one Embodiment of this invention. It is a flowchart which shows the detailed processing content of the production and distribution plan preparation apparatus by another one Embodiment of this invention. It is a figure which shows the example of the macrogene model by one Embodiment of this invention. It is the flowchart which showed the detail of the solution process in a macro solution part. It is a figure which shows the example of the solution result with respect to the macrogene model in one Embodiment of this invention. It is the flowchart which showed the detail of the solution search process in a micro solution part.

Explanation of symbols

20 Condition setting and taking-in part 21 Incoming tank plan making part 22 Gantt chart display / raw material storage tank stock transition graph display part 23 Operator evaluation part 24 Pro-con 25 Vidicon 301 Input data taking part 302 Plan creation period setting part 303 Setting of plan creation time accuracy Unit 304 instruction calculation period setting unit 305 plan plan determination period setting unit 306 extraction unit 307 construction unit 308 macro mathematical model construction unit 309 macro solution unit 310 micro mathematical model construction unit 311 micro solution unit 312 simulation unit 313 decision unit 314 judgment unit 315 Update unit 1001 Input data capture unit 1002 Plan creation period setting unit 1003 Plan creation time accuracy setting unit 1004 Instruction calculation period setting unit 1005 Plan plan decision period setting unit 1006 Extraction unit 1007 Construction unit 1008 Macro gene Dell construction unit 1009 macro solution finding unit 1010 micro priority model construction unit 1011 micro-solving unit 1012 simulation unit 1013 decision unit 1014 judging section 1015 update unit

Claims (21)

In a production / distribution plan creation device for creating a production / distribution plan in a manufacturing process / conveyance that processes or conveys multiple products in different multi-process routes or conveyance routes,
A simulator that simulates the manufacturing process / convey that expresses the logistics state and logistics constraints of the manufacturing process / conveyance,
The optimization process (macro optimization process) is performed for the instruction calculation period (macro instruction calculation period) set in advance from the production start date and time of the production / distribution plan. Macro planning device for calculating)
Given a macro logistics instruction obtained by the macro planning device, an optimization process (micro optimization) is performed for an instruction calculation period (micro instruction calculation period) shorter than a macro instruction calculation period set in advance from the planning start date and time. A production / distribution plan creation apparatus comprising a micro-planning apparatus that performs processing) and calculates a distribution instruction (micro-distribution instruction) for the simulator,
(A) First, from the planning start date and time, the macro planning device creates a macro logistics instruction for the macro instruction calculation period,
(B) Given the macro logistics instruction, the logistics instruction is calculated for the micro instruction calculation period by the micro optimization process by the micro planning device, and is given to the simulator for the preset simulation period. Run the simulation,
(C) The simulation result is confirmed as a production / distribution plan for a predetermined micro-plan decision period,
(D) The date and time immediately after confirmation is set as the new planning start date and time, and the simulation is executed based on the new logistics instruction newly obtained by the micro-optimization process, given the production and logistics plan that has already been established. , Perform a series of processes to finalize the production and logistics plans for the new micro-plan finalization period for each new planning start date and time,
(E) When the production / distribution plan for the macro plan confirmation period is confirmed by the above-described series of processing, a new planning start date and time is set, and the already established production / distribution plan is given. A production / distribution plan creation device that creates a production / distribution plan for a desired plan creation period by performing a series of processes consisting of steps (i) to (e).   All or any of the macro instruction calculation period, the micro instruction calculation period, the simulation period, the macro plan determination period, and the micro plan determination period can be individually set for each repetition loop of the series of processes. The production / distribution plan creation device according to claim 1. The macro planning device constructs a mathematical model (macro mathematical model) composed of the logistics state and logistics constraints of the manufacturing process / transport based on the preset macro time accuracy,
3. The production / distribution plan creation apparatus according to claim 1, wherein macro optimization processing is performed using the macro mathematical model.   4. The production / distribution plan creation device according to claim 1, wherein the macro planning device performs macro optimization processing using a preset evaluation function (macro evaluation function). The micro-planning device builds a mathematical model (micro-mathematical model) consisting of the logistics state and logistics constraints of the manufacturing process / transport based on the preset micro time accuracy,
5. The production / distribution plan creation apparatus according to claim 1, wherein a micro optimization process is performed using the micro mathematical model.   6. The production / distribution plan creation device according to claim 1, wherein the micro planning device performs micro optimization processing using a preset evaluation function (micro evaluation function).   7. The system according to claim 3, wherein all or any of the time accuracy of the macro mathematical model and the micro mathematical model can be individually set for each repetition loop of the series of processes. The production / logistics planning device described.   The production / distribution plan creation apparatus according to claim 1, wherein the macro optimization process performs an optimization calculation process as an optimization or quasi-optimization problem.   9. The production / distribution plan creation apparatus according to claim 1, wherein the micro optimization process performs an optimization calculation process as an optimization or quasi-optimization problem.   In one or both of the macro mathematical model and the micro mathematical model, an overflow prevention variable that stores a value overflowing from the upper limit constraint is introduced for the variable for which the upper limit constraint is set, and the introduced overflow prevention variable is used as the upper limit constraint. 10. The production / distribution plan creation device according to claim 3, wherein a value added to the upper limit value is changed to a constraint in which a variable for which the upper limit constraint is set is equal.   In one or both of the above macro formula model and micro formula model, an interrupt compensation variable that stores a value for interrupting the lower limit constraint to a variable for which a lower limit constraint is set is introduced, and the introduced interrupt compensation variable is set to the lower limit The production / distribution plan creation device according to any one of claims 3 to 10, wherein a value subtracted from a lower limit value of the constraint is changed to a constraint in which the variable set with the lower limit constraint is equal. .   12. The production / distribution plan creation device according to claim 4, wherein one or both of the macro evaluation function and the micro evaluation function include both or one of an overflow prevention variable and an interrupt compensation variable. .   13. The production / distribution plan creation apparatus according to claim 1, wherein a part of the production / distribution plan can be designated in advance when the production / distribution plan is created. A production / distribution plan creation method for creating a production / distribution plan in a manufacturing process / conveyance in which a plurality of products are processed or conveyed by different multi-process routes or conveyance routes,
A simulation process that simulates the manufacturing process / transport expressing the logistics state and logistics constraints of the manufacturing process / transport,
Based on the preset time accuracy (macro time accuracy) for the preset target period (macro instruction calculation period) from the start date of the production / distribution plan, the logistics state of the manufacturing process / transport A macro formula model construction process for building a formula model (macro formula model) composed of distribution constraints and
A macro formula model optimization step of setting an evaluation function (macro evaluation function) for evaluating the macro formula model and performing an optimization process using the macro formula model and the macro evaluation function;
Given a physical distribution instruction (macro physical distribution instruction) obtained by the macro mathematical model optimization process, a target period (micro instruction calculation period) shorter than a macro instruction calculation period set in advance from the planning start date and time Based on the set accuracy (micro time accuracy), a micro formula model construction process for constructing a formula model (micro formula model) composed of the logistics state and logistics constraints of the manufacturing process / transport,
A micro formula that sets an evaluation function (micro evaluation function) for evaluating the micro formula model, performs an optimization process using the micro formula model and the micro evaluation function, and calculates a physical distribution instruction for the simulator A model optimization process,
(A) First, from the planning start date and time, a macro logistics instruction for the macro instruction calculation period is created by the macro mathematical model construction process, the macro evaluation function setting process, and the macro mathematical model optimization process,
(B) Given the macro logistics instruction, calculate the logistics instruction for the micro instruction calculation period by the micro mathematical model optimization process, and execute the simulation for the preset simulation period in the simulation process. And
(C) The result of the simulation process is determined as a production / distribution plan for a predetermined plan determination period (micro plan determination period) (production / distribution plan determination process),
(D) The date and time immediately after confirmation is set as a new planning start date and time, a new micro mathematical model is constructed based on the production / logistics plan that has already been confirmed, and is obtained by a micro optimization process for the micro mathematical model. Update process (micro update process) that executes a simulation based on the new logistics instructions and performs a series of processes to finalize the production / logistics plan for the new microplan decision period for each new planning start date and time )When,
(E) a macro execution determination step for determining whether a predetermined plan determination period (macro plan determination period) has been determined by the micro update step;
Further, when the macro plan confirmation period is confirmed, a new planning start date and time is set, and given the production / distribution plan already determined by the production / distribution plan determination step, the above (a) to (e The production / distribution plan creation method is characterized in that a production / distribution plan for a desired plan creation period is created by a series of steps consisting of the steps (1). A distribution control device for controlling physical distribution in a manufacturing process / conveyance in which a plurality of products are processed or conveyed by different multi-step routes or conveyance routes,
A simulator that simulates the manufacturing process / convey that expresses the logistics state and logistics constraints of the manufacturing process / conveyance,
The optimization process (macro optimization process) is performed for the instruction calculation period (macro instruction calculation period) set in advance from the production start date and time of the production / distribution plan. Macro planning device for calculating)
Given a macro logistics instruction obtained by the macro planning device, an optimization process (micro optimization) is performed for an instruction calculation period (micro instruction calculation period) shorter than a macro instruction calculation period set in advance from the planning start date and time. A logistics control device comprising a micro-planning device that performs processing) and calculates logistics instructions (micro logistics instructions) for the simulator,
(A) First, from the planning start date and time, the macro planning device creates a macro logistics instruction for the macro instruction calculation period,
(B) Given the macro logistics instruction, the logistics instruction is calculated for the micro instruction calculation period by the micro optimization process by the micro planning device, and is given to the simulator for the preset simulation period. Run the simulation,
(C) The simulation result is confirmed as a production / distribution plan for a predetermined micro-plan decision period,
(D) The date and time immediately after confirmation is set as the new planning start date and time, and the simulation is executed based on the new logistics instruction newly obtained by the micro-optimization process, given the production and logistics plan that has already been established. , Perform a series of processes to finalize the production and logistics plans for the new micro-plan finalization period for each new planning start date and time,
(E) When the production / distribution plan for the macro plan confirmation period is confirmed by the above-described series of processing, a new planning start date and time is set, and the already established production / distribution plan is given. ) To (e) to perform a series of processes to create a production / distribution plan for a desired planning period,
A physical distribution control apparatus that controls physical distribution in the manufacturing process and the process to be transported based on the produced production / distribution plan. A distribution control method for controlling distribution in a manufacturing process / conveyance in which a plurality of products are processed or conveyed by different multi-step routes or conveyance routes,
A simulation process that simulates the manufacturing process / transport expressing the logistics state and logistics constraints of the manufacturing process / transport,
Based on the preset time accuracy (macro time accuracy) for the preset target period (macro instruction calculation period) from the start date of the production / distribution plan, the logistics state of the manufacturing process / transport A macro formula model construction process for building a formula model (macro formula model) composed of distribution constraints and
A macro formula model optimization step of setting an evaluation function (macro evaluation function) for evaluating the macro formula model and performing an optimization process using the macro formula model and the macro evaluation function;
Given a physical distribution instruction (macro physical distribution instruction) obtained by the macro mathematical model optimization process, a target period (micro instruction calculation period) shorter than a macro instruction calculation period set in advance from the planning start date and time Based on the set accuracy (micro time accuracy), a micro formula model construction process for constructing a formula model (micro formula model) composed of the logistics state and logistics constraints of the manufacturing process / transport,
A micro formula that sets an evaluation function (micro evaluation function) for evaluating the micro formula model, performs an optimization process using the micro formula model and the micro evaluation function, and calculates a physical distribution instruction for the simulator A model optimization process,
(A) First, from the planning start date and time, a macro logistics instruction for the macro instruction calculation period is created by the macro mathematical model construction process, the macro evaluation function setting process, and the macro mathematical model optimization process,
(B) Given the macro logistics instruction, calculate the logistics instruction for the micro instruction calculation period by the micro mathematical model optimization process, and execute the simulation for the preset simulation period in the simulation process. And
(C) The result of the simulation process is determined as a production / distribution plan for a predetermined plan determination period (micro plan determination period) (production / distribution plan determination process),
(D) The date and time immediately after confirmation is set as a new planning start date and time, a new micro mathematical model is constructed based on the production / logistics plan that has already been confirmed, and obtained by micro optimization processing for the micro mathematical model. Update process (micro update process) that executes a simulation based on the new logistics instructions and performs a series of processes to finalize the production / logistics plan for the new microplan decision period for each new planning start date and time )When,
(E) a macro execution determination step for determining whether a predetermined plan determination period (macro plan determination period) has been determined by the micro update step;
Further, when the macro plan confirmation period is confirmed, a new planning start date and time is set, and given the production / distribution plan already determined by the production / distribution plan determination step, the above (a) to (e ) To create a production / distribution plan for the desired plan creation period through a series of steps consisting of
A logistics control method characterized by controlling logistics in the manufacturing process and the process to be transported based on the created production / distribution plan.   The production / distribution plan creation device according to claim 1, wherein the macro planning device performs a semi-optimization process using a metaheuristic method.   18. The production / distribution plan creation apparatus according to claim 1, wherein the micro-planning apparatus performs a semi-optimization process using a metaheuristic technique.   A program for causing a computer to execute the production / distribution plan creation method according to claim 14.   A program for causing a computer to execute the physical distribution control method according to claim 16.   A computer-readable recording medium on which the program according to claim 19 or 20 is recorded.

Images