WO2022176073A1

WO2022176073A1 - Production planning support system

Info

Publication number: WO2022176073A1
Application number: PCT/JP2021/005976
Authority: WO
Inventors: 明岸田; 知章掛田
Original assignee: 株式会社日立製作所
Priority date: 2021-02-17
Filing date: 2021-02-17
Publication date: 2022-08-25

Abstract

This production planning support system (200) has a step delay analyzing unit (111) that analyzes conditions under which delays in the various steps of producing a product and delays in delivery occur and calculates an expected value of a conditional delay probability and number of delay days for each production condition relating to the steps to generate analyzed information, and a production plan creation/optimization unit (113) that creates a plurality of production plan candidates for the product, calculates, relating to all the steps constituting the production plan candidates, a delivery delay probability and number of delivery delay days of each product in the product plan candidates on the basis of the expected value of the conditional delay probability and the number of delay days based on the analyzed information, adds the delivery delay probability and the number of delay days of the products to calculate a delivery delay assessment index relating to the production plan candidates, and, among the product plan candidates, adopts the product plan candidates having the lowest delivery delay assessment index.

Description

Production planning support system

The present invention relates to a production planning support system that generates information for performing production planning consisting of multiple steps related to design, procurement, and manufacturing processes.

The present invention relates to planning production plans related to procurement and manufacturing from the design process in the manufacturing industry. Existing systems for planning production include Material Requirement Step S Planning (MRP) and Advanced Planning and Step Scheduling (AP Step S). Consider resource procurement plans and production plans. However, these mechanisms are not premised on delays in the processes and procured parts that are part of the production plan.

On the other hand, in Patent Document 1, the entire process is based on the cost of procuring ahead of the delivery date on the standard schedule and the cost of procuring after the delivery date on the standard schedule. A production planning system is disclosed that minimizes the expected cost of production.

In addition, in Patent Document 2, by performing statistical processing from the manufacturing performance information of the process, the probability distribution of the required time for each component is calculated, the probability of delay with respect to the delivery date of the component is calculated, and the manufacturing delay is predicted at the manufacturing order stage. A delivery date management support system is disclosed.

JP 2012-181626 A JP-A-2005-071136

However, the production management system disclosed in Patent Document 1 aims for optimization by comparing the costs associated with delaying procured items and the costs associated with bringing forward the delivery date of procured items. For example, if there are resource constraints due to delays in processes related to in-house production, it is not possible to deal with cases where moving forward a process leads to a shortage of resources for another process.

In addition, although the delivery date management support system disclosed in Patent Document 2 can predict process delays, it does not formulate production plans based on the prediction results.

The present invention is an invention for solving the above-mentioned problems, and is a production planning support system that can realize reliable customer delivery by creating a production plan with a low risk of delivery delay in product production planning. intended to provide

In order to achieve the above object, the production planning support system of the present invention includes order information, procurement information, constraint information, resource information, warehouse information, supplier information, bill of materials information, past production plan information, past production results. Based on the information in the database containing information, analyze the conditions that cause delays in each process of the product and delays in delivery, and calculate the expected value of the conditional delay probability and delay days for each production condition related to the process. a process delay analysis unit as analysis information; a plurality of candidates for the production plan of the product; and the expected value of the number of days of delay, calculate the expected value of the probability of delay in delivery and the number of days of delay in delivery for each product in the production plan candidate, and add up the expected value of the probability of delay in delivery and the number of days of delay in delivery of each product. and a production plan creating unit that calculates a delivery delay evaluation index for the production plan candidate, and adopts a production plan candidate having a small delivery delay evaluation index among the production plan candidates. Characterized by Other aspects of the present invention are described in embodiments below.

According to the present invention, it is possible to achieve reliable customer delivery by creating a production plan with low delivery delay risk in the production plan of the product.

1 is a diagram showing the basic configuration of a production planning support system according to an embodiment; FIG. 4 is a flowchart showing production plan calculation processing according to the first embodiment; It is a figure which shows the example of the past production plan information. It is a figure which shows the example of past production performance information. It is a figure which shows the example of the analysis information which the process delay analysis part analyzed. It is a figure which shows the example of order reception information. It is a figure which shows the example of constraint condition information. It is a figure which shows the example of component information. It is a figure which shows the example of warehouse information. FIG. 4 is a diagram showing an example of production plan candidate 1 first created by a production plan creation/optimization unit; FIG. 10 is a diagram showing a calculation example of conditional delay probability and expected value of delay days for each process; FIG. 10 is a diagram showing a calculation example of a conditional delay probability and an expected value of delay days for each delivery destination; FIG. 10 is a diagram showing an example of a delivery delay evaluation index for each delivery destination and an overall delivery delay evaluation index; FIG. 10 is a diagram showing a delivery delay probability weighting factor and an expected delivery delay days value weighting factor, which are weighting parameters for the delivery delay probability and expected delivery delay days value set for each delivery destination, which are set in the evaluation parameter input section; . This is an example of production plan candidates calculated by the production plan creation/optimization department. It is an example of another candidate for the production plan calculated by the production plan creation/optimization department. It is an example of another candidate for the production plan calculated by the production plan creation/optimization department. FIG. 11 is a flow chart showing production plan calculation processing in consideration of rescheduling when a delivery delay occurs, according to the second embodiment; FIG. It is a figure which shows the example of order reception information. It is a figure which shows the example of customer information. It is a figure which shows the example of procurement information. FIG. 4 is a diagram showing an example of delay prediction information for each process related to work stored in analysis information; FIG. 10 is a diagram showing an example of delay prediction information of each supplier regarding parts stored in analysis information; FIG. 4 is a diagram showing an initial production plan calculated by a production plan creation/optimization unit; FIG. 18B is a diagram showing an example of a delayed process list in the production plan of FIG. 18A; It is a figure which shows the result of having registered the production plan of FIG. 18A in the production plan list|wrist. FIG. 19B is a diagram showing the results of rescheduling indexes related to the production plan of FIG. 19A; FIG. 22 is a diagram showing a result of additionally registering candidates for the new production plan in FIG. 21; It is a figure which shows the example of the production plan in the precondition of process delay. FIG. 18B is a diagram showing an example of a new production plan candidate different from that in FIG. 18A , in which only the delivery timing of the part 3 of the product 3 is different. 21. It is a figure which shows the example which implemented the review of the production plan of FIG. It is a figure which shows the result of the production risk visualization part which displayed the warning on the display part. 14 is a flow chart showing production plan calculation processing according to the third embodiment.

Embodiments for carrying out the present invention will be described in detail with reference to the drawings as appropriate.
<Embodiment 1>
FIG. 1 is a diagram showing an example of the basic configuration of a production planning support system 200 according to an embodiment of the invention. The production planning support system 200 has a storage unit 100 (database), a processing unit 120, a communication I/F 130, a display unit 140, and an input unit 150, and is connected to the network NW via the communication I/F 130. The storage unit 100 contains order information 101, production plan information 102, actual production information 103, procurement information 104, constraint information 105, resource information 106, warehouse information 107, supplier information 108, bill of materials information 109, analysis information 110. , a production plan list 117, a delayed process list 118, and the like are stored. The processing unit 120 includes a process delay analysis unit 111, a process delay prediction unit 112, a production plan creation/optimization unit 113, an external input/output unit 114, a production plan risk visualization unit 115, an evaluation parameter input unit 116, and the like. Note that the process delay prediction unit 112 and the production plan creation/optimization unit 113 may be combined into a production plan creation unit.

(Storage unit 100 (database))
The external input/output unit 114 stores various information necessary for production planning in the storage unit 100 . Order information 101 stores order information from customers. The procurement information 104 stores procurement information such as which part was ordered from which supplier and when it was delivered. The constraint information 105 stores constraint information regarding each process, such as the precedence-successor relationship between processes, and the resources and locations such as which machines and people are used by the processes. The resource information 106 stores information about how many resources such as people and machines required for production are held. The warehouse information 107 stores the size of the space in the warehouse, information on the weight that can be loaded, and information on the performance and schedule of parts and products stored in the warehouse. The supplier information 108 stores information about suppliers who place orders for parts. The parts list information 109 stores information on the parts that make up each product and the process for manufacturing the parts.

The production plan information 102 stores information on the production plan created by the production plan creation/optimization unit 113 . The actual production information 103 stores information on the actual production performed based on the production plan. The analysis information 110 stores information analyzed by the process delay analysis unit 111 . The production plan list 117 and the delayed process list 118 will be described later.

(Processing unit 120)
The process delay analysis unit 111 collects order information 101, procurement information 104, constraint information 105, resource information 106, supplier information 108, and parts list information 109, and compares the production plan information 102 with the actual production information 103. , statistically analyze the conditions that cause delays in each process and delivery delays, and calculate the expected value of the conditional delay probability and delay days for each production condition related to processes such as products, business partners, parts, and manufacturing quantities, Information is stored in analysis information 110 .

In Patent Document 2, the delay probability for the delivery date of parts is used. On the other hand, in the present embodiment, by setting the delay probability with conditions related to the production conditions, even in the same process of the same product, the delay probability and the delay time may vary depending on the difference in workers and the difference in machinery and equipment. Allowing for changing expectations.

The process delay prediction unit 112 uses information about a process and production conditions related to the process, such as products, suppliers, parts, and production quantities, as input values to calculate a conditional delay probability and an expected value of delay days that match the production conditions. , from the analysis information 110 . The process delay prediction unit 112 passes the acquired expected value of the delay probability with conditions and the number of delay days to the production planning/optimization unit 113 as a return value.

The production plan creation/optimization unit 113 determines when and how to execute which process from order information 101, procurement information 104, constraint information 105, resource information 106, supplier information 108, and parts list information 109. Create production plan candidates. As for the production plan candidates, each production process is assembled so that each product ordered by the customer in the order information 101 meets the delivery date.

The production plan creation/optimization unit 113 inputs information on each process and production conditions for each process to the process delay prediction unit 112 for all the processes that constitute the created production plan candidate, and determines the conditional delay of the process. Receive expected values for probabilities and days late. Based on the expected value of the conditional delay probability and the number of delay days for each process, the production planning/optimization unit 113 calculates the predicted value of the delivery delay probability and the number of days of delay for each product in the production plan candidate. .

The production plan creation/optimization unit 113 acquires evaluation weighting parameters for the delivery delay probability and the delivery delay days from the evaluation parameter input unit 116 . If the contract with the customer regarding damages for late delivery is determined by whether or not the delivery is delayed, this weighting parameter should be set to a large value for the weighting parameter regarding the probability of delivery delay, and the damage according to the number of days of delivery delay If it is money, the weighting parameter for the evaluation of the number of days of delivery delay is set to a large value, so that the difference in contract conditions for each customer can be reflected in the delivery delay evaluation index. In addition, the difference in the amount of damage caused by delayed delivery for each customer can be dealt with by adjusting the value of the weighting parameter for each customer.

The production plan creation/optimization unit 113 multiplies and totals the weighting parameters by the delivery delay probability and delivery delay days of each product to calculate the delivery delay evaluation index for the production plan candidate.

Next, the production plan creation/optimization unit 113 creates candidates for a different production plan under the same input information conditions, and calculates the delivery delay evaluation index by the same method as before. The delivery delay evaluation index of the previous production plan candidate and the new production plan candidate are compared, and the production plan candidate with the smaller delivery delay evaluation index is adopted. By repeating this operation until there are no production plan candidates, the production plan candidate with the smallest delivery delay evaluation index is adopted as the optimum production plan. The production plan creation/optimization unit 113 records the optimized production plan in the production plan information 102, and outputs information on the production plan from the external input/output unit 114 to the outside.

As described above, the external input/output unit 114 stores various information necessary for the production plan in the storage unit 100, and outputs the information on the production plan to the outside via the network NW.

The production planning risk visualization unit 115 visualizes the process on the screen of the display unit 140 so that a process with a large delivery delay evaluation index and a process with a small delivery delay evaluation index can be distinguished.

The evaluation parameter input unit 116 acquires, via the input unit 150, evaluation weighting parameters for the delivery delay probability and the delivery delay days, as described above.

Details of the processing of the production planning support system 200 will be described below.
FIG. 2 is a flow chart showing the production plan calculation process S100 according to the first embodiment. Flowcharts are described with reference to FIGS. 3-12D.

In step S101, the process delay analysis unit 111 collects each information stored in the storage unit 100 shown in FIG. Analytical information 110 is generated which consists of expected values of delay probability and delay days.

FIG. 3 is an excerpt of an example of past production plan information 102. Figure 4. It is an excerpt of an example of past production performance information 103. FIG. The manufacturing number in FIGS. 3 and 4 is an abbreviation for manufacturing number, which is an individual identification number for each product. FIG. 5 is a diagram showing an example of analysis information 110 analyzed by the process delay analysis unit 111. As shown in FIG. In this example, the process delay analysis unit 111 sets production conditions for each worker, product, part, and process, and records and stores in the analysis information 110 the conditional delay probability and the expected delay days value for each production condition. there is According to FIG. 5, for worker A, product 1, and process 2, it can be seen that the delay probability and the expected value of delay days are not "0".

Here, as described above, in this embodiment, the conditional delay probability and the expected number of delay days are adopted. By using conditional delay probabilities related to production conditions, it is possible to take into account that even for the same process of the same product, the expected delay probability and delay time will change due to differences in workers and machinery. 3 and 4, in the case of the past production plan information 102 shown in FIG. , September 6th, and September 8th. However, according to the past production record information 103 in FIG. 4, a delay occurred and one day was added on September 9th. In this case, in the process of worker A, product 1, and process 2 in FIG. 5, the delay probability is 1/4 times=0.25, and the expected number of delay days is 1 day/4 days=0.25. On the other hand, the worker B, the product 1, and the process 2 have four days of September 2nd, September 4th, September 6th, and September 8th. According to it, no delay occurred. In this case, in the process of worker B, product 1, and process 2 in FIG. 5, the delay probability is 0 and the expected number of delay days is 0.

Returning to FIG. 2, in step S102, the production plan creation/optimization unit 113, from each input information of the received order information 101, the constraint information 105, the resource information 106, the warehouse information 107, the supplier information 108, and the parts list information 109, Create a production plan.

FIG. 6 is an example of the order information 101. It stores the order date received from the customer, the scheduled delivery date, the production start possible date, and the production completion deadline date. FIG. 7 shows an example of the constraint information 105, which defines order constraints between processes when producing the product 1, workers who can be in charge of production, standard man-hours, and information on machinery and equipment to be used. In FIG. 7, from the "Preceding Process" column, Process 2 is performed for Product 1 after Process 1, from the "Worker" column, either Worker A or Worker B can handle From the column, it can be seen that the standard man-hour is 1 day, and from the "Machine Equipment" column, no machinery is used.

FIG. 8 is a diagram showing the parts table information 109, which indicates whether each product is produced from parts and processes. In FIG. 1 is composed of part 1, and it can be seen from the "process" column that product 1 is produced from process 1 and process 2, which require part 1.

　Fig. 9 is a diagram showing the warehouse information 107, which consists of the stock quantity of each part and the empty space indicating how many spaces there are for each part on each shelf in each warehouse. As shown in FIG. 9, since the inventory of part 1 exists, in producing the product 1, the part 1 can be allocated from the inventory without placing an order with the supplier.

FIG. 10A shows candidates 1 (production plans P11 and P11A) of the production plan first created by the production plan creation/optimization unit 113 based on the information stored in the storage unit 100 shown in FIG. Worker A plans to carry out Process 1 on October 2nd and Process 2 on October 3rd for products destined for Company A, while for products destined for Company B, Process 1 is scheduled to be carried out by worker B on October 3, and process 2 is to be carried out by worker B on October 4. It can be seen that this production plan candidate 1 maintains consistency with the information in FIGS. For ease of viewing, unlike FIG. 3, FIG. 10A shows the order of production steps in chronological order for each product.

Returning to FIG. 2, in step S103, the production plan creation/optimization unit 113 inputs the information of each process in the production plan candidate of FIG. Get the conditional delay probability and expected days delay. If the delivery destination in FIG. 10A is company A, product 1, process 1, and worker A, both the conditional delay probability of the process and the expected value of delay days are 0 from FIG. Similarly, FIG. 10B (production plan P11B) is obtained by calculating conditional delay probabilities and delay days expected values for other processes from FIG.

In step S104, the production planning/optimization unit 113 calculates the delivery delay probability and the expected value of the number of days of delay for each product from the conditional delay probability and the expected value of the number of days of delay of each process. FIG. 10C (production plan P11C) describes the calculation result. From FIG. 10B, regarding the product for which the delivery destination is Company A, the only process in which the conditional delay probability of the process is not 0 is process 2, and there is no subsequent process to process 2. Therefore, the delivery delay probability and the number of delivery delay days The expected value is 0.25 for each. On the other hand, regarding the product to be delivered to company B, the conditional delay probability of each process is 0, so the delivery delay probability and the expected value of delivery delay days are also 0. Details of FIG. 10D ((production plan P11D) will be described later.

Subsequently, the production planning/optimization unit 113 calculates a delivery delay evaluation index from the delivery delay probability, the expected value of the number of days of delivery delay, and the weighting coefficient of the probability of delivery delay and the expected value of the number of days of delivery delay.

FIG. 11 shows a weighting parameter 119 for the delivery delay probability and the expected delivery delay days set for each delivery destination set in the evaluation parameter input unit 116, which is a delivery delay probability weighting coefficient and a delivery delay days expected value weighting. FIG. 4 is a diagram showing coefficients; In the example of FIG. 11, the same value is set for both company A and company B, which are delivery destinations, but it can be set for each delivery destination.

In FIG. 11, the process delay probability weighting factor and the process delay days expected value weighting factor are weighting factors when calculating the evaluation index of the process delay risk, and the process delay warning criterion warns a process with high delay risk. Parameters for display will be described later.

The delivery delay evaluation index introduced in this embodiment will be described.
A delivery delay evaluation index of a certain product for a certain delivery destination is calculated by the formula (1).
Delivery delay evaluation index = delivery delay probability weighting coefficient × product delivery delay probability + expected delivery delay days weighting coefficient × expected delivery delay days of product (1)

Further, the overall delivery delay evaluation index related to the production plan candidate is calculated by the formula (2) as the total value of the delivery delay evaluation indices of all the products to be produced.
Overall delivery delay evaluation index = Σ delivery delay evaluation index _{for all products to be produced} (2)

In the example of FIG. 11, the delivery delay probability weighting coefficient and the delivery delay days expected value weighting coefficient are all 0.5, and the delivery delay probability and delivery delay days expected value for each delivery destination are the values shown in FIG. The delivery delay evaluation index for each destination and the overall delivery delay evaluation index have values shown in FIG. 10D (production plan P11D).

Returning to FIG. 2, in step S105, the production plan creation/optimization unit 113 determines whether or not there is another production plan candidate. If there is no candidate that satisfies (step S105, No), candidate 1 is regarded as the optimum production plan, and the process ends.

If another production plan candidate exists (step S105, Yes), the production plan creation/optimization unit 113 sets this new candidate as candidate 2 in step S106. The production plan creation/optimization unit 113 generates another production plan candidate by changing the process sequence of the production plan candidate 1 created in step S102.

In steps S106, S107, and S108, the production plan creation/optimization unit 113 executes the same processes as in steps S102, S103, and S104 for candidate 2 of the production plan. Calculate the delay metrics.

In step S109, the production plan creation/optimization unit 113 compares the overall delivery delay evaluation index of the

production plan candidates

1 and 2, and resets the production plan candidate with the smaller value as the production plan candidate 1. , the process returns to step S105.

By repeatedly executing steps S105 to S109 until there are no production plan candidates, the production plan with the smallest overall delivery delay evaluation index is selected.

FIG. 12A is an example of a production plan candidate (production plan P12) calculated by the production plan creation/optimization unit 113. FIG. FIG. 12B is an example of another candidate (production plan P13) for the production plan calculated by the production plan creation/optimization unit 113 . FIG. 12C is an example of another production plan candidate (production plan P14) calculated by the production plan creation/optimization unit 113 .

FIGS. 12A, 12B, and 12C are production plan candidates in this embodiment calculated by the production plan creation/optimization unit 113 in the processing from step S105 to step S109. 12C, which has the smallest overall delivery delay evaluation index, is selected as the production plan from among the production plans of FIGS. 12A to 12C and the production plan candidates of FIG.

Through the above procedures, the production planning support system 200 can create a production plan that minimizes the risk of product delivery delays.

In this embodiment, in step S103 of FIG. 2, the information of the process in the production plan candidate is input to the process delay prediction unit 112, and the conditional delay probability and the expected number of delay days of the process matching the input information are calculated. is obtained, a machine learning method such as a neural network may be used to obtain the conditional delay probability and the expected value of the number of delay days for a process similar to the input information.

<Embodiment 2>
In the second embodiment, when there are multiple production plan process sequences that minimize the risk of product delivery delays created based on the method of the first embodiment, a production plan that can be rescheduled when a delivery delay occurs is selected. can do.

FIG. 13 is a flowchart showing a production plan calculation process S200 that takes rescheduling into account when a delivery delay occurs, according to the second embodiment. This flow chart will be described in detail below with reference to FIGS. 7 to 9 and 14 to 23. FIG.

FIG. 14 is a diagram showing an example of the order information 101. FIG. The scheduled delivery dates for Product 2, Product 3, and Product 4 are November 5th. Constraint condition information 105 regarding product 2, product 3, and product 4 is shown in FIG. 7, parts table information 109 is shown in FIG. 8, and warehouse information 107 is shown in FIG.

FIG. 15 is a diagram showing an example of supplier information 108. FIG. The supplier information 108 describes the supplier of the parts, the handling parts, the address of the supplier, and the like. FIG. 16 is a diagram showing an example of the procurement information 104. As shown in FIG. The procurement information 104 describes the order dates, delivery schedules, and the like of parts 2, 3, and 4. FIG.

Products 2, 3, and 4 of the order information 101 in FIG. 14 require

parts

1, 2, 3, and 4 from the parts list information 109 in FIG. Also, from the warehouse information 107 in FIG. 9, it can be seen that since

parts

1 and 2 are in stock, they can be used for production, and part 3 is out of stock. The fact that the part 2 is not shown in FIG. 9 means that it is a part that cannot be stored in a warehouse, that is, a part that needs to be manufactured immediately after delivery. Therefore, in this embodiment, it is necessary to order the parts 2 and 3 in order to produce the product.

Returning to FIG. 13, in step S201, the processing unit 120 performs each process from step S101 to step S109 in the production plan calculation process shown in FIG. create a production plan that

FIG. 17A is a diagram showing an example of delay prediction information 110a for each process related to work stored in analysis information 110. FIG. As shown in FIG. 17A, each condition related to work includes combinations of workers, machinery, products, parts, and processes, and the delay probability and the expected delay days value for each combination are calculated. For example, in the case of "worker A, machine C, product 2, part 2, process 1", the delay probability is 0 and the expected number of delay days is 0, indicating that no delay will occur.

FIG. 17B is a diagram showing an example of delay prediction information 110b for each supplier regarding parts stored in analysis information 110. FIG. FIG. 17B was calculated by the process delay analysis unit 111 by statistically analyzing the procurement information 104 of FIG. In FIG. 17B, in the case of part 2 and company X, the delay probability is 0.5 and the delay days expected value is 0.5. is 2/4=0.5, and since there is a delay of 2 days with respect to the delivery date of 4 days in total, the expected value of delay days is 2/4=0.5.

FIG. 18A is a diagram showing the initial production plan P21 calculated by the production plan creation/optimization unit 113. FIG. FIG. 18B is a diagram showing an example of the delayed process list 118 in the production plan of FIG. 18A. Here, a method for minimizing the delivery delay risk of the product based on the production plan P21 in FIG. 18A and the delay prediction information in FIGS. 17A and 17B will be described. In FIG. 18A, the date on which the part is registered means the timing when the part was delivered and became available for production.

Returning to FIG. 13, in the process of step S202, the production plan and the overall delivery delay evaluation index calculated in step S201 are registered in the production plan list 117. FIG. 19A is a diagram showing the result (production plan list 117a) of registering the production plan P21 of FIG. 18A in the production plan list 117. FIG. At the time of processing in step S202, the initial value of 0 is set for the rescheduling index in FIG. 19A because the value is updated as a result of the subsequent processing.

Returning to FIG. 13, in the process of step S203, for each process of the production plan newly registered in the production plan list 117, from the analysis information 110, the process whose conditional delay probability is not 0 is extracted, and the delayed process Stored in list 118 along with the conditional delay probability and expected delay days.

Fig. 18A shows the delivery delay probability, the expected value of the number of days of delivery delay, etc., but does not show the probability of delay and the expected value of the number of days of delay for each process. The probabilities of delay and expected days of delay for each step are determined in view of FIGS. 17A and 17B. From the delay prediction information 110a of each process related to the work in FIG. 17A, all the conditional delay probabilities that the work of each process will be delayed are 0. On the other hand, from the delay prediction information 110b of each supplier related to the parts of FIG. has a probability of 0.5 that is delayed, there is a risk that only process 1 of product 2 using part 2 will be delayed.

The production planning/optimization unit 113 registers delay processes with conditional delay probabilities other than 0 and related information for identifying the processes in the delay process list 118 . FIG. 18B is the delayed process list 118 in the production plan of FIG. 18A, and stores information on delivery destinations, products, and dates as information for specifying delayed processes. It also stores the conditional delay probability that the process will be delayed, the expected value of the number of delay days, and the No. which means the line number.

Returning to FIG. 13, in the processes of steps S204 and S205, the production plan creation/optimization unit 113 selects one process from the delay process list 118, and assumes that it is delayed by the number of delay days expected value, and then prepares the production plan. recreate. In this embodiment, as shown in FIG. 18B, only Process 1 of Product 2 with Company C as the delivery destination is registered in the delayed process list 118, so this process is set to be delayed by one day. Here, in FIG. 18B, the expected number of days of delay is 0.5 days, but in this embodiment, since the minimum unit of the process is 1 day, the number after the decimal point is rounded up and the number of days of delay is set to 1 day. .

In the process of step S205, the production plan creation/optimization unit 113 assumes the process delay of step S204, that is, the delivery destination is Company C, the process 1 of the product 2 is delayed by one day, and the process is delayed on November 3rd. On the premise that it has been completed, the processing from step S102 to step S109 is executed to calculate the production plan and the overall delivery delay evaluation index. FIG. 20 shows the calculation result of step S205 in this precondition.

Compare the initial production plan P21 in FIG. 18A with the production plan P22 calculated in step S205 in FIG. In the case of production plan P22 in FIG. 20, company C/product 2 is scheduled for November 3 and November 4 with a one-day delay, and company D/product 3 is delayed by one day due to worker A's relationship. For Company E, Product 4, Process 2 is scheduled for November 5 with a shift of one day due to Worker A's relationship. As a result, the delivery delay probability of company D/product 3 and company E/product 4 is 1, and the expected value of the number of delivery delay days is 1. As a result, the delivery delay evaluation index becomes 1, and the overall delivery delay evaluation index becomes 2.0. In addition, the thick-line frame in the table of FIG. 20 indicates the changed part from FIG. 18A.

In the process of step S206, the production plan creation/optimization unit 113 calculates the updated value of the rescheduling index from the following formula (3) and adds it to the rescheduling index in the production plan list. Here, the reason why the conditional delay probability of the process selected in step S204 is multiplied in the equation (3) is to consider the likelihood of delay occurring in the process of step S204.

Updated value of rescheduling index=Overall delivery delay evaluation index calculated in step S205×Conditional delay probability of the process selected in step S204 (3)

In this embodiment, the overall delivery delay evaluation index is 2.0 from FIG. 20, and the conditional delay probability of the process is 0.5 from FIG. 18B, so the updated rescheduling index is 1.0. . This value is added to the rescheduling index of the production plan registered in the production plan list 117 in step S202. FIG. 19B shows the result (production plan list 117b) of adding the updated value of the rescheduling index to the rescheduling index.

In the process of step S207, the production planning/optimization unit 113 determines whether or not there is an unselected process in the delayed process list. If there is an unselected process (step S207, Yes), the process returns to step S204. If there is no unselected process (step S207, No), the process proceeds to step S208. Here, as shown in FIG. 18B, only one process is delayed, so the process proceeds to step S208. In this embodiment, only one process is registered in the delayed process list, but if there are multiple processes, the updated value of the rescheduling index for each process is added to the rescheduling index. Become.

In the process of step S208, the production plan creation/optimization unit 113 selects a production plan that is not registered in the production plan list 117 and that has the same value as the overall delivery delay evaluation index of the production plan calculated in step S201. Explore plans. FIG. 21 shows a candidate for a new production plan P23 different from that in FIG. 18A, in which only the delivery timing of part 3 of product 3 is different from that in FIG. 18A. It should be noted that the bold line frame in the table of FIG. 21 is the changed part.

In the process of step S209, the production plan creation/optimization unit 113 determines whether or not there is a new production plan candidate as a result of the search, and if a new production plan candidate is found (step S209, Yes), The process proceeds to step S210. On the other hand, when there is no new production plan candidate (step S209, No), the production plan creation/optimization unit 113 proceeds to step S211.

In the process of step S210, the production plan creation/optimization unit 113 additionally registers the new production plan candidate found in step S208 in the production plan list 117, and then returns to the process of step S203. FIG. 19C is a diagram showing the result (production plan list 117c) of additionally registering candidates for the new production plan P23 of FIG. As in step S202, the rescheduling index of the new production plan No. 2 is set to 0, which is the initial value.

The production plan creation/optimization unit 113 performs the processes from step S203 to step S207 for the new production plan candidate in step S208. The processing result of step S203 is the same as in FIG. 18A, and the processing result of steps S204 and S205 for the new production plan candidate of step S208 is shown in production plan P24 of FIG. As shown in FIG. 21, the candidate for the new production plan P23 in step S208 is that the delivery date of part 3 is one day earlier than that in FIG. be. Therefore, in the production plan P24 of FIG. 22, the overall delivery delay evaluation index is 0 when process 1 is delayed.

That is, the production plan P23 of FIG. 21 is compared with the production plan P24 of the calculation result of step S205 of FIG. In the case of the production plan P24 in FIG. 22, Company C/Product 2 is scheduled for November 3 and November 4 with a one-day delay, and Company D/Product 3 is scheduled for November 2nd by worker A. Therefore, the schedule is shifted one day earlier to November 2nd and November 3rd, and Company E's product 4 is scheduled in the same way as in FIG. As a result, the delivery delay probability of company C/product 2, company D/product 3, and company E/product 4 is 0, and the expected value of days of delivery delay is 0. As a result, the delivery delay evaluation index is set to 0, and the overall delivery delay evaluation index is set to 0. In addition, the thick-line frames in the table of FIG. 22 indicate the changed parts from FIG.

The updated value of the rescheduling index is the conditional delay probability of the process from FIG. When the calculation of step S206 is performed, the value is 0. Therefore, as a result of performing the processing from step S203 to step S207, the value of the rescheduling index of the candidate for the production plan P23 in FIG. 21 remains 0, and the value of the production plan list is not updated from FIG. 19C.

Then, the process proceeds to steps S208 and S209. In this embodiment, since there is no other production plan candidate, the process proceeds to step S211 in the determination process of step S209.

In the process of step S211, the production plan creation/optimization unit 113 selects the production plan with the smallest rescheduling index from the production plan list 117. As a result of performing a series of processes from step S201 to step S211, the production plan list 117 is shown in FIG. 19C, so the production plan No. 2 in FIG. is calculated as a possible production plan.

From the analysis information 110 and the production plan list 117, the production plan risk visualization unit 115 displays a warning on a display device such as a display for processes that are greater than the process delay warning standard specified in the evaluation parameter input unit 116. Note that the process delay warning criterion is calculated from the following equation (4).

Process delay warning criteria = process delay probability weighting coefficient x process delay probability + expected number of days of process delay weighting factor x expected number of days of process delay (4)

FIG. 23 is a diagram showing the result (screen 141) of the warning displayed on the display unit by the production risk visualization unit. In this embodiment, based on the setting values of the evaluation parameter input section 116 of FIG. 10 and the analysis information of FIG. 17, the process 1 of the product 2 is displayed as a warning with a solid line frame as a process with a high possibility of being delayed. . In addition, when the process 1 is delayed, the process that can be replaced is displayed with a dotted line frame. With such a display, the production planning support system 200 makes it easy for the user to understand which process has a high delay risk and how to deal with the delay. In addition, the delivery delay probability, the expected value of delivery delay days, the delivery delay evaluation index, and the overall delivery delay evaluation index are changed from the information in the production plan list 117 to the values when the processes are rearranged, that is, the values shown in FIG. is doing.

With the above procedure, the production planning support system 200 according to the present embodiment can calculate a production plan that can be rescheduled when a delivery delay occurs. Provides functions for grasping and responding to processes.

<Embodiment 3>
In the production planning calculation methods based on

Embodiments

1 and 2, calculations are performed for all combinations of process sequences, combinations of resources, etc. Therefore, as the number of products and processes increases, the amount of calculation can be very large and the computation time can be very long. In the third embodiment, production plans for a plurality of products are originally created at the same time. , is characterized by creating a production plan with a low risk of delay in delivery.

FIG. 24 is a flowchart showing production plan calculation processing S300 according to the third embodiment.
In step S301, the process delay analysis unit 111 collects each piece of information stored in the storage unit 100 shown in FIG. .

In step S<b>302 , the production plan creation/optimization unit 113 creates a production plan for each product to be produced, and calculates the delivery delay probability of each product from the analysis information 110 .
In the process of step S303, the production plan creation/optimization unit 113 sorts the products in descending order of the delivery delay probability.

In the process of step S304, the production plan creation/optimization unit 113 creates a production plan for the product with the highest delivery delay probability under the conditions of the input information. When creating the production plan first, the product with the highest delivery delay probability is created under conditions that allow free use of resources and process schedules, so the production plan with the lowest delivery delay probability is created. be able to.

Subsequently, in the process of step S305, the production plan creation/optimization unit 113 checks whether the upper limit for creating the production plan for each product set by the external input/output unit 114 has been reached. If not (step S305, No), the process proceeds to step S306.

In the process of step S306, the production plan creation/optimization unit 113 creates a production plan for the product with the second highest probability of delivery delay within the range of resources and process schedules not used in step S304. After creating the production plan, the production plan creation/optimization unit 113 returns to the process of step S305, and repeats the process of step S306 as long as the upper limit for creating the production plan for each product is not exceeded.

When the upper limit for creating a production plan for a single product has been exceeded by repeated processing (step S305, Yes), the production plan creation/optimization unit 113 advances the process to step S307. In step S307, the production plan creation/optimization unit 113 performs the following operations for all remaining products for which production plans have not been created, under the conditions of resources and process schedules not used in steps S304 to S306. The processes from step S102 to step S109 are executed to create a production plan.

By limiting the upper limit for creating a production plan for a single product to the number that can be executed by a computer, it is possible to create a production plan with a low risk of delivery delays while reducing the calculation time.

In this embodiment, the probability of delay in delivery and the number of days (probability of delay in delivery, expected number of days in delay in delivery) are used as evaluation values, but the risk of delay is evaluated in consideration of the penalty amount for delay in delivery. may

The production planning support system 200 of the present embodiment includes history information related to past production plans for design work, procurement work, and manufacturing work; Constraints on production, including the dependency relationship that indicates the preceding-succeeding relationship between each process and the upper limit number of human resources and machinery that can be used at a certain time, and the history of process delays in part or all of each process, For the storage unit 100, which stores the history of delivery delays of products produced in each process, history information related to production plans accumulated in the storage unit 100 and information related to process delays or parts delivery delays from suppliers A process delay analysis unit 111 that calculates the conditional delay probability and the expected value of the delay time for each production condition in each process from history information, and the execution order and execution timing of each process based on the constraint conditions. Recombination, each process in this production plan, the expected value of delay probability and delay time with conditions that match or are similar to each production condition are obtained from history information, and the probability of product delivery delay occurring from these values A process delay prediction unit 112 that calculates the expected value of the number of delivery delay days, an evaluation parameter input unit 116 that sets a weighting parameter for a delivery delay evaluation index related to the delivery date, a weighting parameter, the probability of product delivery delay occurring, and delivery a production planning/optimizing unit 113 that calculates a delivery delay evaluation index from the expected value of the delay days and creates a process sequence that minimizes the delivery delay evaluation index from among the combinations of selectable process sequences; have.

According to this embodiment, in production planning, the risk of process delay occurrence is predicted from history information of past process delays, and while observing resource constraints such as facilities and personnel and execution order constraints between processes, Based on the delay risk, it is possible to formulate a production plan to minimize the risk of product delivery delay and to consider rescheduling when a delay occurs.

The process delay analysis unit 111 analyzes process delays such as differences in parts suppliers, customers, part types, etc., differences in design personnel, manufacturing personnel, and procurement personnel, differences in manufacturing order, design of order quantity, and part delivery. occurrence probability and delay time can be evaluated.

When creating a production plan, the production plan creation/optimization unit 113 refers to the delay evaluation result described above, calculates a delivery delay prediction in the created production plan, and selects the most delayed delivery from among various production plan candidates. A low-risk production plan can be selected and output.

According to the production planning support system 200 of the present embodiment, it is possible to achieve reliable customer delivery by creating a production plan with a low risk of delay in delivery in the production planning of products.

100 storage unit (database)
101 Order information 102 Production plan information 103 Actual production information 104 Procurement information 105 Constraint information 106 Resource information 107 Warehouse information 108 Supplier information 109 Parts list information 110 Analysis information 110a Delay forecast information for each process 110b Delay forecast information for each supplier 111 process delay analysis unit 112 process delay prediction unit (production plan creation unit)
113 Production Planning and Optimization Department (Production Planning Department)
114 external input/output unit 115 production planning risk visualization unit 116 evaluation parameter input unit 117 production planning list 118 delay process list 120 processing unit 130 communication I/F
140 display unit 150 input unit 200 production planning support system NW network S100, S200, S300 production planning calculation processing

Claims

Based on database information including order information, procurement information, constraint information, resource information, warehouse information, supplier information, parts list information, past production plan information, and past production performance information, each process of the product A process delay analysis unit that analyzes the conditions that cause delays and delivery delays, calculates the expected value of the conditional delay probability and delay days for each production condition related to the process, and uses it as analysis information;
A plurality of production plan candidates for the product are created, and for all the processes that constitute the production plan candidates, the production is performed based on the conditional delay probability of each process based on the analysis information and the expected value of the number of delay days. By calculating the expected value of the delivery delay probability and the delivery delay days of each product in the plan candidate, and adding the delivery delay probability and the delivery delay days expected value of each product, the delivery delay related to the production plan candidate A production planning support system, comprising: a production planning unit that calculates an evaluation index and adopts a production planning candidate with a small delivery delay evaluation index from among the production planning candidates.
When there are a plurality of candidates for the process order that minimize the delivery delay evaluation index, the production planning unit simulates the delay of the process whose delay probability is not 0, and simulates the process that is simulated to cause a delay. 2. The production planning support system according to claim 1, wherein the possibility of replacing a process with another process is evaluated, and a process sequence is created so that a process having a delay probability and another process can be replaced.
2. The production planning support system according to claim 1, wherein the production planning unit preferentially allocates the production plan for the product with the high delivery delay probability to the production plan.
The production planning support system further comprises
an evaluation parameter input unit for inputting an evaluation weighting factor for the delivery delay probability of the product and the expected value of delivery delay days;
2. The production planning support system according to claim 1, wherein the delivery delay evaluation index is calculated by formula (1).
Formula (1)
Delivery delay evaluation index = delivery delay probability weighting coefficient × product delivery delay probability + expected delivery delay days weighting coefficient × expected delivery delay days of product
History information related to production plans for past design, procurement, and manufacturing operations; information related to human resources and machinery required in each process of design, procurement, and manufacturing; Constraints on production, including relationships and upper limits on the number of human resources and machinery that can be used at any given time, history of process delays in some or all of the above-mentioned processes, and the number of products produced by each of the above processes For a storage unit in which the history of delivery delays is accumulated, each process of production is processed based on the history information on the production plan accumulated in the storage unit and the history information on process delays or parts delivery delays from business partners. A process delay analysis unit that calculates the conditional delay probability and the expected value of the delay time for each production condition in
Based on the constraint conditions, the execution order and execution timing of each process are rearranged, and each process and each production condition in this production plan are matched or similar, and the expected value of delay probability and delay time with conditions is obtained from the history information. a process delay prediction unit that obtains and calculates the expected value of the probability of product delivery delays and the number of delivery delay days from these values;
an evaluation parameter input unit for setting a weighting parameter for a delivery delay evaluation index related to the delivery date;
A delivery delay evaluation index is calculated from the weighting parameter, the probability of occurrence of the delivery delay of the product, and the expected value of the delivery delay days, and the delivery delay evaluation index is selected from among the combinations of selectable process orders. A production planning support system characterized by having a production planning/optimization section that creates a process sequence such that: