JP2010001114A - Vessel service schedule preparation method and device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a vessel service schedule preparation method and a device preparing the service schedule of the vessel determining the optimum call order for transportation from an amount of an article transported from a plurality of vessels (convoy) for a predetermined period and the number of voyages of its course and simultaneously making attaining of the number of voyages in every determined course and equalization of a lifting place arrival interval compatible. <P>SOLUTION: The vessel service schedule preparation method has the optimum voyage number determination step of determining the call order and the number of voyages while taking into consideration to a restriction condition not depending on a time including at least a loadable amount of the vessel and the number of vessels from a loading place, a lifting place and a transportation amount of a transportation object; and the optimum service schedule preparation step of preparing the service schedule from the call order and the number of voyages determined by the optimum voyage number determination step. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a ship operation plan creation method and apparatus for optimally determining a port calling order, the number of voyages, and a cargo handling timing when a predetermined amount of luggage is delivered by a plurality of ships.

  Shipment by ship is a major factor in the transportation of industrial products and semi-finished products, and it is important to efficiently carry out this operation plan and management. In order to operate efficiently, it is necessary to have plan management that comprehensively considers the operation of loading / unloading sites such as berth plans as well as ship operation plans. However, although there is a mechanism for optimizing these individually, it is still a difficult task to create and manage an efficient operation plan as a whole, and it is left to the judgment of skilled workers There are many current situations.

  As an effort to solve these problems, for example, in Patent Document 1, a plurality of operation plan drafts are created on a simulation basis based on a predetermined call order selection rule for a specific ship, and the most criterion is among them. Techniques for adopting adapted plans are disclosed.

  In addition, Patent Document 2 covers a relatively short period of time as an integer planning problem that minimizes costs by appropriately selecting a voyage route under the inventory restriction of loading and landing sites. A technique for creating an operation plan by modeling and sequentially connecting optimum solutions at each time point is disclosed.

  Patent Document 3 discloses a technique for correcting a long-term operation plan problem of a ship, a ship assignment plan provisionally created by a person in charge of ship assignment in advance, and correcting the voyage schedule of a candidate ship for plan correction within an allowable range. Is disclosed.

  Further, Patent Document 4 discloses a technique for creating an operation plan including a ship by a linear programming method.

In addition to the above-mentioned patent documents, non-patent documents referred to in the disclosure of the invention are also described below.
JP-A-5-12308 JP 2000-172745 A JP 2006-111387 A JP 2000-57220 A "Let's use mixed integer programming (MIP): modeling practice and its background", Mathematical Modeling Laboratory, Naoji Nosue, ILOG Optimization seminar, http://www.math-model.co.jp/ILOG_OPT_Seminer_2006_Rev .pdf K. Nonobe, and T. Ibaraki, "Formulation and tabu search algorithm for the resource constrained project scheduling problem", Essays and Surveys in Metaheuristics, Kluwer Academic Publishers, pp.557-588, 2002

  Ship operation planning problems are classified into NP (Non-deterministic Polynomial) difficult problems that can be solved in real time by applying mathematical optimization techniques in one step. . In other words, since it is necessary to satisfy various constraints over all the times in the plan creation period, the more the number of ships and luggage to be targeted, or the longer the plan creation period, the more the solution will be obtained. Becomes difficult.

  However, since the technique disclosed in Patent Document 1 is a method for preliminarily setting a port calling order selection rule that is supposed to give an optimal solution empirically, and creating and evaluating a plurality of operation plan proposals on a simulation basis. There is a problem that there is no guarantee about optimality.

  In addition, the technology disclosed in Patent Document 2 attempts to optimize costs comprehensively by dividing the plan creation target period into short periods and connecting the operations considered to be optimal in each period. Therefore, due to the voyage decided earlier, there is a possibility that it will cause excess or shortage of inventory at the loading / unloading site and competition / standby of ships. In addition, there is a problem that formulation as an integer programming problem becomes complicated, such as handling time of loading / unloading work and waiting time when there are multiple ships competing.

  In addition, the technique disclosed in Patent Document 3 is a method of correcting a provisional plan created in advance by a ship allocator within a certain allowable range without creating an operation plan from scratch. Therefore, the operation plan is greatly affected by the provisional plan and the allowable range, and the resulting operation plan is not necessarily optimal. Also, it goes without saying that it is not useful for a new operation plan where an operation plan must be created from scratch. At present, veteran ship dispatchers are making provisional plans, and it is clear that they should not be influenced by the skills of the staff.

  Furthermore, although the technique disclosed in Patent Document 4 can optimize the number of ships for each load capacity to be used, it does not use the concept of time, so it creates an operation plan along the time axis. There is a problem that it cannot be done.

  In general, in an operation plan based on an optimization method, it is difficult to incorporate some conditions required for actual operation, for example, equidistant departure / arrival. This is because the goal is contrary to efficiency.

  The present invention has been made in view of such a problem, and obtains the optimal port calling order (route) and the number of voyages of the route from the quantity transported by a plurality of vessels (fleet) in a predetermined period, An object of the present invention is to provide a ship operation plan creation method and apparatus for creating a ship operation plan that simultaneously achieves the required number of voyages for each route and equalizes landing landing arrival intervals at the same time. To do.

  The invention according to claim 1 of the present invention is a ship operation plan creation method for planning operations in a predetermined period of a plurality of ships that share a loading area or a landing area, and includes a loading area, a landing area, and a transportation object for transportation. The optimal number of voyages and the number of voyages in consideration of constraints that do not depend on time, including at least the amount of ships that can be loaded and the number of vessels, and the voyage order determined in the optimum number of voyages And a ship operation plan creation method characterized by having an optimum operation plan creation step of creating an operation plan from the number of voyages.

  The invention according to claim 2 of the present invention is the ship operation plan creation method according to claim 1, wherein in the optimum voyage number determination step, a preset route candidate is input and the number of voyages is limited. This is a method for creating a ship operation plan.

  Furthermore, the invention according to claim 3 of the present invention is a ship operation plan creation device that plans operation in a predetermined period of a plurality of ships that share a loading place or a landing place, and a loading place, a landing place, The optimum voyage number determination processing means for obtaining the calling order and the number of voyages in consideration of the constraints not depending on the time including at least the loadable amount of ships and the number of ships from the transport amount, and the optimum voyage number determination means It is a ship operation plan creation device characterized by having an optimum operation plan creation processing means for creating an operation plan from a port call order and the number of voyages.

  According to the present invention, in the optimum number of voyage determination process, the loadable capacity of the ship, the operation rate, the number of ships, the possibility of calling at a port (cannot enter a port unless a specific ship is used), the cargo handling capacity per time (of the landing site) In order to solve the optimization problem with the upper and lower limits depending on the crane capacity and storage capacity) as a constraint, the order of port calls and the number of voyages can be optimized regardless of the skill of the person in charge. In addition, time-dependent constraints such as cargo handling competition at the landing site are not taken into consideration, but only time-independent constraints such as load capacity and the number of ships that can be used are considered, and time-dependent constraints are integrated into the optimal operation plan creation process. By doing so, the number of variables to be optimized can be reduced, and an optimum solution can be obtained in a short time. Moreover, by dividing the preparation of the operation plan at this stage, it is possible to make adjustments such as eliminating the appearance of a port with a large number of cargo handling operations per operation.

  In order to realize the optimal number of voyages and the optimal number of voyages determined in the optimal voyage number determination process (and adjustment by the person in charge), the optimal operation plan preparation process, such as the inspection of vessels and port equipment during the planning period, etc. Determine the cargo handling time at each port of call in consideration of non-operating hours, and create an optimal operation plan. At this stage, since the scheduling target job (hereinafter referred to as a voyage job) to be arranged on the time axis has already been found, it is possible to apply various types of scheduling specialized algorithms.

  In addition, because the order of port calls for the optimal number of voyage determination process is given in advance in consideration of past results (delay of midway voyage, possibility of competition for cargo handling at the landing site), the permutation and combination of the target ports Compared to the case of automatic generation, it is possible to eliminate an unrealistic calling order in advance and to obtain an optimal solution in a shorter time.

  The present invention optimizes an operation plan for the entire plan creation period at once, so that the entire operation plan can be created by connecting a conventional method for creating an operation plan by simulation or an operation plan that seems to be optimal at that time. There is no possibility of causing competition or waiting of the vessel at the loading / unloading site later due to the predetermined voyage that can occur in the method of creation.

  Hereinafter, the present invention will be specifically described with reference to the drawings and mathematical expressions. FIG. 1 is a diagram showing an overall processing flow of a ship operation plan creation method according to the present invention. The main processes are performed in the order of the optimum voyage number determination step (Step 1), the correction step (Step 2), and the optimum operation plan creation step (Step 3). In the optimal voyage number determination process (Step 1), the ships that are subject to the operation plan are treated as the same, such as loading capacity, navigation speed, etc. Eventually, in the optimum operation plan creation process (Step 3), a process of handling each ship individually is performed.

  In FIG. 1, first, data such as a loading place, a landing place, a type and a transportation amount (the number of packages) of a transport object are input, and data necessary for creating a plan is read from a master database or a work database.

  The route candidate database, which is one of the data to be read, is a port call order (also referred to as a route) that is automatically generated or registered manually, for example, as shown in Table 1 below.

  Each route is given a unique route code, and the required number of ports of call are listed in the order of port calls. The route is associated with a group of a plurality of ships (hereinafter referred to as a hull form) having the same attributes such as the loadable capacity and speed. Therefore, even if the calling order is the same, if the ship type is different, the route will be different. Since the maximum and minimum number of voyages immediately after loading the route candidate database is determined only from the equipment capacity and operating conditions, this is corrected to the constraints for the preparation of the current operation plan. Change.

  Hereinafter, the optimal voyage number determination step (Step 1), the correction step (Step 2), and the optimal operation plan creation step (Step 3) will be described in detail in this order.

  First, in the optimal number-of-voyage determination step (Step 1), the optimal port calling order (route) and the number of voyages are determined from the loading area, loading location, and the amount of transportation (number of luggage) of the given transportation object (package type). The problem is formulated as a mixed integer programming and solved to find the optimal port order (route) and the number of voyages.

  The symbols used in the formulation are summarized below. First, variables are defined as shown in (1) below.

  Various constants are defined as shown below.

  Further, using the above symbols, the constraint conditions (constraints 1 to 9) are formulated as follows.

  The above constraint conditions are examples, and additions and deletions are appropriately performed according to the target, and the following optimization calculation is performed. In the optimization calculation, for example, the following evaluation function (evaluation 1 to 4) is selected and optimized according to the problem.

  In the above formulation, since variables representing time are not included, the number of variables to be optimized is extremely small, and an optimum value can be calculated within a practical time. Further, the route candidate may be created by any method. For example, a permutation of the target port of call may be automatically generated or selected from routes having a track record.

  By solving the optimization problem formulated as described above in the optimal voyage number determination step (Step 1), the number of voyages for each route and the number of cargoes for each type of cargo at each port of call, where the evaluation function is the optimum value And the number of cargo handling and cargo handling time are obtained. Although there are various solutions here, for example, the solution disclosed in Non-Patent Document 1 may be used.

  Next, in the correction step (Step 2), necessary correction is added to the output result of Step 1. Necessary correction here is a condition that is difficult to set as a constraint condition for optimization calculation. It is difficult to make a decision with automatic calculation, but it is not preferable because it is inefficient but not human. It is a correction by the judgment. As a correction method, for example, route candidates are added / changed, various upper / lower limit values included in the formulation of the optimum voyage number determination step (Step 1) are corrected, and the optimum voyage number determination step (Step 1) is executed again. The method of doing is taken. The result of completion of the correction is passed to the optimum operation plan creation step (Step 3) as a voyage job to be executed. Here, the voyage job is data including loading time, voyage time, unloading time, waiting time of each work, and the like.

  In the optimal operation plan creation process (Step 3), the navigation job passed from the correction process (Step 2) is arranged on the time axis while solving various conflicts, and an actual operation plan is created. That is, as the input data, the voyage job, the operating time and holiday data of the port of call, the non-operating time zone such as dock / inspection of individual ships, the target completion date and time of each voyage, and the optimal scheduling to assign this to multiple ships The problem will be solved.

  This problem is a general scheduling problem, and various efficient solutions can be applied. For example, the solution disclosed in Non-Patent Document 2 may be used. In addition, with regard to the equality of departure / arrival required for actual navigation, it is possible to provide an interval from the preceding voyage job as a prior relationship between the voyage jobs.

  An embodiment in which the present invention is applied to an actual ship operation plan is shown below. FIG. 2 is a diagram illustrating a device configuration example according to the present embodiment. In the figure, reference numerals are assigned to the main ones, 1 is a master database, 2 is a work database, 3 is an optimum voyage number determination processing means, and 4 is an optimum operation plan creation processing means.

  Input data such as loading place, landing place, luggage type, and number of luggage is input from the input means and sent to the optimum voyage number determination processing means 3.

  The master database 1 includes a ship master, a port master, and a navigation time master, and is stored in data storage means. The ship master records the ship type, the number of ships, the ship name, the loadable load type, and the loadable load number for each individual ship. For each berth, the port master records the port name, number of berths, ship types that can be called, holidays, dredging, cargo type and cargo handling time. The navigation time master records the departure port, arrival port, navigation distance, and navigation time by ship type for all combinations of ports.

  The work database 2 includes a route candidate database, and the data storage means stores the order of calling, the type of cargo handling cargo, the number of cargo handling cargo, the time required for cargo handling, and the time required for navigation. Only the order of port calls is included at the time of the start of the operation plan. As the result of the optimal voyage number determination process (Step 1) in the optimal voyage number determination processing means 3, the cargo type, the number of cargoes, the required time for loading, The time required for navigation is written. When waiting time data is added to this, it becomes voyage job data.

  Then, as the evaluation function, select one of the above-mentioned maximizing the amount of transported luggage, maximizing the number of voyages, maximizing (the number of luggage × transport distance), and minimizing the total required time, and determining the optimum voyage number It is sent to the means 3 to start the optimum voyage number determining step (Step 1).

  Next, we will explain the specific operation plan creation procedure and contents. First, the loading area, landing area, type, and transportation amount (number of packages) of the object to be transported are given as input data as shown in Table 2 below, and “(number of packages x transport distance) maximization” is set as the evaluation function. When the optimum voyage number determination step (Step 1) is selected, the results shown in Tables 3 and 4 are output. The route candidate database shown in Table 1 above was used.

  Table 3 shows how many cargoes of the loading / unloading / loading type are allocated to which route, as a result, how many voyages each route should take, and how many cycle times at that time, etc. Output is shown. Table 4 shows the number of port calls at each port in this case. In this example (Table 3), the number of packages carried on the Nl_02 route is 0, indicating that it is not necessary to use this route (see Table 3).

  If for some reason it is necessary to transport luggage on this route (Nl_02), the minimum number of voyages of Nl_02, which is a candidate route in Table 1, is not 0 in the correction process (Step 2) of the correction processing means. The optimum number of voyages may be set again, and the optimum number of voyages determination step (Step 1) in the optimum number of voyages determination processing means 3 may be executed again.

  Tables 5 and 6 show the results when the minimum number of voyages of the route code Nl_02 shown in Table 1 is changed from 0 to 2 and re-executed. Comparing this result with Tables 3 and 4, the two cruises of Nl_03 are simply transferred to Nl_02. When both results are evaluated in more detail, both the Nl_02 route and the Nl_03 route include the route from port a2 through port b1 to port b5, but not at port b1 (loading site a2 It is clear that there is no baggage at the landing b1). That is, it can be seen that it is not necessary to call at port b1 on the way from port a2 to port b5 for both routes.

  Therefore, in the correction process (Step 2) of the correction processing means, route candidates are added / changed until such inadequacy disappears (for example, the calling order of a2−b1−b5 for the calling order of Nl_02 and Nl_03) , A2−b5 calling order), correction of various upper and lower limit values included in the formulation of the optimum voyage number determination step (Step 1), and execution of the optimum voyage number determination step (Step 1) again. repeat.

  In this way, it is difficult to find the operation plan creation problem when executing the final operation plan at once by dividing the process so that the correction process is performed at the stage when the optimal voyage determination process is completed. It is also possible to detect and correct inappropriateness.

  Next, the route candidate database of the work amount database 2 in which the execution results of the optimum voyage number determination step (Step 1) and the correction step (Step 2) are written, and each master of the master database 1 are read. The route candidate database in which the execution results of the optimum voyage number determination step (Step 1) and the correction step (Step 2) are written is as shown in Table 7 below. Note that Table 7 is also voyage job data because it includes data such as loading time, voyage time, unloading time, and waiting time.

  At this stage, since the route candidate database includes all information of the voyage job, it is converted into an input data format to the scheduling engine built in the optimum operation plan creation means 4 and the scheduling engine is started. This optimum operation plan creation means 4 is internally divided into a data format conversion unit and a scheduling engine unit, and it is possible to know only the scheduling engine, for example, a scheduling engine equipped with another scheduling algorithm.

  The execution result of the scheduling engine is a voyage job optimally arranged on the time axis, and is output in the form of an operation schedule and a Gantt chart. Table 8 shown below is an output example of an operation schedule table of an arbitrary ship, and FIG. 3 is a diagram showing an example of an output Gantt chart.

  In this embodiment, the input means, the master database, the work database, the optimum voyage number determination processing means, the correction processing means, and the optimum operation plan creation processing means 4 may be constituted by one computer. Alternatively, it may be distributed among a plurality of computers.

As described above, according to the operation plan creation method of the present invention, since the operation plan creation is divided into the optimum voyage number determination step and the operation plan creation step, it has the following characteristics. is there.
(1) The order of port calls (routes) and the number of voyages can be optimized regardless of the skill of the person in charge.
(2) Evaluate the results of the optimal voyage number determination process, and eliminate the inappropriate route, giving the person in charge the opportunity to make adjustments.
(3) The result of the optimum voyage number determination process is in the form of a general scheduling problem, and various efficient solutions can be applied.

It is a figure which shows the whole processing flow of the ship operation plan preparation method which concerns on this invention. It is a figure which shows the apparatus structural example in a present Example. It is a figure which shows the example of the output Gantt chart.

Explanation of symbols

1 Master database 2 Work database 3 Optimum voyage number determination processing means 4 Optimal operation plan creation processing means

Claims (3)

A ship operation plan creation method for planning operations for a predetermined period of a plurality of ships sharing a loading or unloading site,
An optimal voyage number determining step for obtaining a port calling order and the number of voyages in consideration of constraints that do not depend on time including at least the loadable amount of vessels and the number of vessels, from the loading place, landing place, and transportation amount of the object to be transported;
A ship operation plan creation method, comprising: an optimum operation plan creation step of creating an operation plan from the port call sequence and the number of voyages obtained in the optimum number of voyages determination step. In the ship operation plan creation method according to claim 1,
In the optimal voyage number determining step, a route candidate set in advance is input and the number of voyages is limited. A ship operation plan creation device that plans operations for a predetermined period of a plurality of ships that share a loading or unloading site,
Optimum voyage number determination processing means for determining the calling order and the number of voyages from the loading place, the landing place of the transportation object, and the transportation amount in consideration of at least constraints that do not depend on the time including the loadable capacity and the number of ships. ,
An apparatus for creating a ship operation plan, comprising: an optimum operation plan creation processing means for creating an operation plan from the port calling sequence and the number of voyages obtained by the optimum number-of-voyages determination means.

Images