Intelligent decision-making system based on reinforcement learning

Abstract

The invention discloses an intelligent decision-making system based on reinforcement learning. The system comprises a production line yield acquisition module, a production line power consumption acquisition module, a production line labor cost accounting module, an input module, an accounting module and an output module. The production line yield collection module can collect yield information of different production lines, the yield information is the quantity of products produced by the production lines every day, and the yield information of the different production lines is marked as A1, A2,..., An; the production line power consumption acquisition module can acquire power consumption information of different production lines, the power consumption information is electric quantity needing to be consumed by the production lines per hour, the power consumption of the different production lines is marked as B1, B2,..., Bn, when the system is implemented, the cost of order completion is evaluated according to the yield, the energy consumption and the labor cost of the different production lines, and the cost of order completion is calculated according to the evaluation result. Production suggestions are provided for enterprises, so that the enterprises are helped to save cost and improve benefits.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an intelligent decision-making system based on reinforcement learning. Background Art

[0002] In the design and production of the factory, different production lines will be designed. During the busy season, there are more orders and multiple production lines will run at the same time to meet production needs. During the off-season, there are fewer orders. After receiving the order, the production line will be selectively opened according to the order quantity and the length of the construction period.

[0003] The output, power consumption and labor costs of different production lines are different. After receiving an order, according to the order volume, a production line with a larger output will generally be selected to start. However, a production line with a larger output not only means an increase in production costs, but also requires more staff to participate in it. However, a production line with a larger output is not necessarily the most in line with production needs. For this reason, this application proposes an intelligent decision-making system based on reinforcement learning. Summary of the invention

[0004] To this end, the present application provides an intelligent decision-making system based on reinforcement learning to solve the problem in the prior art that a suitable production line cannot be selected according to the order quantity.

[0005] In order to achieve the above objectives, this application provides the following technical solutions:

[0006] In a first aspect, an intelligent decision-making system based on reinforcement learning includes a production line output collection module, a production line power consumption collection module, a production line labor cost accounting module, an input module, an accounting module, and an output module;

[0007] The production line output collection module can collect the output information of different production lines, where the output information is the amount of products produced by the production line every day, and mark the output information of different production lines as A1, A2, ..., An;

[0008] The production line power consumption collection module can collect power consumption information of different production lines, the power consumption information is the power consumed by the production line per hour, and the power consumption of different production lines is marked as B1, B2, ..., Bn;

[0009] The production line labor cost accounting module can calculate the labor costs of different production lines and mark the labor costs of different production lines as C1, C2, ..., Cn;

[0010] The input module is used to input order information, which includes the quantity of products required in the order and the time when the order is completed;

[0011] The accounting module, after the input module inputs the order information, can calculate the production method with the lowest production cost based on the information collected by the production line output collection module, the production line power consumption collection module, and the production line labor cost accounting module;

[0012] The output module can output the calculation result after the accounting module calculates the lowest production mode.

[0013] Preferably, when calculating the labor cost of the production line, the labor cost of the production line is calculated based on the number of personnel required when the production line is running and the wages of the personnel.

[0014] Preferably, the calculation module performs calculation according to the following steps:

[0015] Mark different production lines as X1, X2, X3...Xn,

[0016] The time required for production line X1 to complete an order is Y1, the power consumed is P1, and the labor cost is Z1;

[0017] The time required for production line X2 to complete an order is Y2, the power consumed is P2, and the labor cost is Z2;

[0018] The time required for production line X3 to complete the order is Y3, the power consumed is P3, and the labor cost is Z3;

[0019] The time required for production line Xn to complete an order is Yn, the power consumed is Pn, and the labor cost is Zn;

[0020] According to the time required by the order, the production lines that cannot complete production within the specified time will be screened out, and then the production cost of the production line will be calculated based on the power consumption and labor costs of the production line, and the calculation results will be sent to the output module.

[0021] Preferably, after receiving the calculation results of the accounting module, the output module outputs the production line with the lowest cost.

[0022] Preferably, the accounting module has an electricity cost calculation unit, and after the accounting module screens out production lines that do not meet the order requirements, the electricity cost calculation unit calculates the electricity cost according to the power consumption of the production lines.

[0023] Preferably, the accounting module also has a production scheduling unit, which obtains information based on the electricity price and the distribution of different electricity prices at different times of the day. When the production line can be completed according to the order requirements, the production scheduling unit calculates the opening and closing time of the production line based on the electricity price information to complete the production schedule. After the production schedule is completed, the electricity cost calculation unit calculates the electricity cost of the production line based on the electricity price information in the production schedule. When the accounting module sends the calculation result to the output module, the production schedule is synchronously sent to the output module, and the output module can also output the production schedule for the production line.

[0024] Preferably, a production line operation and maintenance cost collection module is also included, and the operation and maintenance cost collection module is used to collect the operation and maintenance cost of the production line and the production line loss cost.

[0025] Preferably, a communication module is also included, and after the output module outputs the result, the communication module can send the output result to a communication terminal.

[0026] Preferably, an intelligent analysis module is also included. When the output module outputs the result, the intelligent analysis module can obtain the output result. A storage unit is provided in the intelligent analysis module, and the storage unit stores the output result. When a new order is input into the system, the intelligent analysis module can compare the new order information with the stored old order information. When the new order information is the same as the old order information, the output module directly outputs the result stored in the storage module.

[0027] Compared with the prior art, this application has at least the following beneficial effects:

[0028] 1. When this system is implemented, the cost of completing orders is evaluated based on the output, energy consumption and labor costs of different production lines, and production suggestions are provided to enterprises, thereby helping enterprises save costs and improve efficiency;

[0029] 2. When this system is implemented, new orders can be compared with old orders. When the orders are the same, the results can be directly output, thereby speeding up the processing speed of this system. At the same time, it can further improve the accuracy of cost accounting, making it easier for users to grasp costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more intuitively illustrate the prior art and the present application, exemplary drawings are given below. It should be understood that the specific shapes and structures shown in the drawings should not generally be regarded as limiting conditions for implementing the present application; for example, those skilled in the art are capable of easily making conventional adjustments or further optimizations to the addition / reduction / attribution division, specific shapes, positional relationships, connection methods, dimensional ratios, etc. of certain units (components) based on the technical concepts and exemplary drawings disclosed in the present application.

[0031] Figure 1 A module diagram of an intelligent decision-making system based on reinforcement learning provided in Example 1 of the present application. DETAILED DESCRIPTION

[0032] The present application is further described below in detail through specific embodiments in conjunction with the accompanying drawings.

[0033] An intelligent decision-making system based on reinforcement learning, comprising a production line output collection module, a production line power consumption collection module, a production line labor cost accounting module, an input module, an accounting module and an output module;

[0034] The production line output collection module can collect the output information of different production lines, where the output information is the amount of products produced by the production line every day, and mark the output information of different production lines as A1, A2, ..., An;

[0035] The production line power consumption collection module can collect power consumption information of different production lines, the power consumption information is the power consumed by the production line per hour, and the power consumption of different production lines is marked as B1, B2, ..., Bn;

[0036] The production line labor cost accounting module can calculate the labor costs of different production lines and mark the labor costs of different production lines as C1, C2, ..., Cn;

[0037] The input module is used to input order information, which includes the quantity of products required in the order and the time when the order is completed;

[0038] The accounting module, after the input module inputs the order information, can calculate the production method with the lowest production cost based on the information collected by the production line output collection module, the production line power consumption collection module, and the production line labor cost accounting module;

[0039] The output module can output the calculation result after the accounting module calculates the lowest production mode.

[0040] When this system is implemented, the cost of completing an order is evaluated based on the output, energy consumption and labor costs of different production lines, and production suggestions are provided to the company, thereby helping the company save costs and improve efficiency.

[0041] When calculating the labor cost of a production line, the labor cost of the production line is calculated based on the number of personnel required when the production line is running and the wages of the personnel.

[0042] The accounting module performs calculations according to the following steps:

[0043] Mark different production lines as X1, X2, X3, ... Xn, that is, the first production line is marked as X1, the second production line is marked as X2, and so on;

[0044] The time required for production line X1 to complete an order is Y1, the power consumed is P1, and the labor cost is Z1;

[0045] The time required for production line X2 to complete an order is Y2, the power consumed is P2, and the labor cost is Z2;

[0046] The time required for production line X3 to complete the order is Y3, the power consumed is P3, and the labor cost is Z3;

[0047] The time required for production line Xn to complete an order is Yn, the power consumed is Pn, and the labor cost is Zn;

[0048] According to the time required by the order, the production lines that cannot complete production within the specified time will be screened out, and then the production cost of the production line will be calculated based on the power consumption and labor costs of the production line, and the calculation results will be sent to the output module.

[0049] In some embodiments: the factory has three production lines, namely X1, X2, and X3, the output A1 of X1 is capable of producing and processing 10 tons of finished products per day, the output A2 of X2 is capable of producing 30 tons of finished products per day, and the output A3 of X3 is capable of producing 50 tons of finished products per day. The order requires 800 tons of finished products and is required to be completed within 40 months.

[0050] In this embodiment, production line X1 takes 80 days to complete the order, which obviously does not meet the requirements of the order, so production line X1 is excluded; production line X2 takes 27 days to complete the order, which meets the requirements of the order; production line X1 takes 16 days to complete the order, which meets the requirements of the order.

[0051] The power consumption of production line X1 is 1000 kWh per hour, the power consumption of production line X2 is 5000 kWh per hour, and the power consumption of production line X3 is 10000 kWh per hour.

[0052] Production line X1 requires 10 workers during production, production line X2 requires 15 workers during production, and production line X3 requires 20 workers during production.

[0053] From the above information, it can be concluded that production line X2 and production line X3 can meet production needs, and the cost of production line X2 is 27 days of electricity cost and the wages of 15 workers, and the cost of production line X3 is 16 days of electricity cost and the wages of 20 workers.

[0054] After calculating the costs of X2 and X3, compare the two costs and select the production line with lower cost.

[0055] After receiving the calculation results of the accounting module, the output module outputs the production line with the lowest cost.

[0056] The accounting module includes an electricity cost calculation unit. After the accounting module screens out production lines that do not meet the order requirements, the electricity cost calculation unit calculates the electricity cost according to the power consumption of the production lines.

[0057] In actual applications, the price of electricity is different every day. When calculating the electricity cost of the production line, the electricity cost calculation unit can calculate the electricity cost more accurately based on the electricity consumption in different time periods.

[0058] The accounting module also has a production scheduling unit, which obtains information based on the electricity price and the distribution of different electricity prices at different times of the day. When the production line can be completed according to the order requirements, the production scheduling unit calculates the opening and closing time of the production line based on the electricity price information to complete the production schedule. After the production schedule is completed, the electricity cost calculation unit calculates the electricity cost of the production line based on the electricity price information in the production schedule. When the accounting module sends the calculation result to the output module, the production schedule is synchronously sent to the output module, and the output module can also output the production schedule for the production line.

[0059] When arranging production tasks, if the construction period is sufficient, in order to save production costs, the production plan of the production line can be planned. That is, production can be carried out during the period when the electricity price is low, and production can be avoided when the electricity price is high or as little as possible when the electricity price is high, thereby better saving costs and improving corporate benefits.

[0060] It also includes a production line operation and maintenance cost collection module, which is used to collect the operation and maintenance costs and production line loss costs of the production line.

[0061] When the production line is in production, there will be not only the above-mentioned costs, but also other maintenance costs, parts replacement costs and other costs. Through the operation and maintenance costs, other costs besides the above-mentioned are counted in order to screen out production lines with lower costs.

[0062] It also includes a communication module. After the output module outputs the result, the communication module can send the output result to a communication terminal.

[0063] After the production line and production schedule are determined, the system can communicate with the staff's mobile phone, and then send the production line to be opened and the production schedule to the staff's mobile phone so that the staff can perform specific operations.

[0064] When enterprises are operating and producing, most of them have fixed customers, and the orders of fixed customers are mostly the same or periodic. In order to enable the production schedule to be exported more quickly, the following technical solution is set up: when the output module outputs the result, the intelligent analysis module can obtain the output result, and a storage unit is provided in the intelligent analysis module, which stores the output result. When a new order is input into the system, the intelligent analysis module can compare the new order information with the stored old order information. When the new order information is the same as the old order information, the output module directly outputs the result stored in the storage module without going through the accounting module for calculation.

[0065] At the same time, after the order is completed, the staff can also compare the costs at each location with the results calculated by this system. When the two are the same, no operation is performed. When the two are different, the actual cost is entered and the storage unit saves the actual cost for subsequent output of the production line cost.

[0066] The technical features of the above embodiments may be arbitrarily combined (as long as there is no contradiction in the combination of these technical features). To make the description concise, not all possible combinations of the technical features in the above embodiments are described; these embodiments that are not explicitly written should also be considered to be within the scope of this specification.

Claims

1. An intelligent decision-making system based on reinforcement learning, characterized in that: It includes production line output collection module, production line power consumption collection module, production line labor cost accounting module, input module, accounting module and output module; The production line output collection module can collect the output information of different production lines, where the output information is the amount of products produced by the production line every day, and mark the output information of different production lines as A1, A2, ..., An; The production line power consumption collection module can collect power consumption information of different production lines, the power consumption information is the power consumed by the production line per hour, and the power consumption of different production lines is marked as B1, B2, ..., Bn; The production line labor cost accounting module can calculate the labor costs of different production lines and mark the labor costs of different production lines as C1, C2, ..., Cn; The input module is used to input order information, which includes the quantity of products required in the order and the time when the order is completed; The accounting module, after the input module inputs the order information, can calculate the production method with the lowest production cost based on the information collected by the production line output collection module, the production line power consumption collection module, and the production line labor cost accounting module; The output module can output the calculation result after the accounting module calculates the lowest production method.

2. The intelligent decision-making system based on reinforcement learning according to claim 1, characterized in that: When calculating the labor cost of a production line, the labor cost of the production line is calculated based on the number of personnel required when the production line is running and the wages of the personnel.

3. The intelligent decision-making system based on reinforcement learning according to claim 1, characterized in that: The accounting module performs calculations according to the following steps: Mark different production lines as X1, X2, X3, ... Xn; The time required for production line X1 to complete an order is Y1, the power consumed is P1, and the labor cost is Z1; The time required for production line X2 to complete the order is Y2, the power consumed is P2, and the labor cost is Z2; The time required for production line X3 to complete the order is Y3, the power consumed is P3, and the labor cost is Z3; The time required for production line Xn to complete an order is Yn, the power consumed is Pn, and the labor cost is Zn; According to the time required by the order, the production lines that cannot complete production within the specified time will be screened out, and then the production cost of the production line will be calculated based on the power consumption and labor costs of the production line, and the calculation results will be sent to the output module.

4. The intelligent decision-making system based on reinforcement learning according to claim 3, characterized in that: After receiving the calculation results of the accounting module, the output module outputs the production line with the lowest cost.

5. The intelligent decision-making system based on reinforcement learning according to claim 4, characterized in that: The accounting module includes an electricity cost calculation unit. After the accounting module screens out production lines that do not meet the order requirements, the electricity cost calculation unit calculates the electricity cost according to the power consumption of the production lines.

6. The intelligent decision-making system based on reinforcement learning according to claim 5, characterized in that: The accounting module also has a production scheduling unit, which obtains information based on the electricity price and the distribution of different electricity prices at different times of the day. When the production line can be completed according to the order requirements, the production scheduling unit calculates the opening and closing time of the production line based on the electricity price information to complete the production schedule. After the production schedule is completed, the electricity cost calculation unit calculates the electricity cost of the production line based on the electricity price information in the production schedule. When the accounting module sends the calculation result to the output module, the production schedule is synchronously sent to the output module, and the output module can also output the production schedule for the production line.

7. The intelligent decision-making system based on reinforcement learning according to claim 5, characterized in that: It also includes a production line operation and maintenance cost collection module, which is used to collect the operation and maintenance costs and production line loss costs of the production line.

8. The intelligent decision-making system based on reinforcement learning according to claim 5, characterized in that: It also includes a communication module. After the output module outputs the result, the communication module can send the output result to a communication terminal.

9. The intelligent decision-making system based on reinforcement learning according to claim 1, characterized in that: It also includes an intelligent analysis module. When the output module outputs the result, the intelligent analysis module can obtain the output result. A storage unit is provided in the intelligent analysis module, and the storage unit stores the output result. When a new order is input into the system, the intelligent analysis module can compare the new order information with the stored old order information. When the new order information is the same as the old order information, the output module directly outputs the result stored in the storage module.

Images