Weighted order real-time batch processing system and method for resource conflicts

Abstract

The application provides a system and method for real-time batch processing of orders with weights for resource conflicts, and belongs to the technical field of intelligent manufacturing and computer control. The system comprises an order receiving module, a working area monitoring module, an order information processing module, a job unit selection module and a job unit construction module. The order receiving module updates the order information arriving at the system in real time; the working area monitoring module updates the job unit state in the working area in real time; the order information processing module generates a to-be-constructed job unit according to the to-be-processed order information; the job unit selection module selects a job unit for construction in the next batch; and the job unit construction module determines whether to immediately construct the selected job unit for construction and decides the construction sequence and construction time of the job unit. The application can effectively face diversified, personalized, sharply fluctuating and rapidly changing demands, quickly process orders with weights, and improve the rapid response capability of the system.

Description

Technical Field

[0001] This invention relates to a real-time batch processing system and method for orders, specifically a real-time batch processing system and method for weighted orders to address resource conflicts, belonging to the field of intelligent manufacturing and computer control technology. Background Technology

[0002] Group technology is an intelligent management technology for multi-variety, small-batch production. It is widely used in flexible manufacturing systems, effectively improving the ability of industrial automation systems to be flexibly reconfigured. Extensive practical production experience and research results show that deploying operations in a unit-based manner can effectively improve production quality, reduce operating costs, increase work efficiency, and shorten operation cycles. In recent decades, with the rapid development of science and technology and the rapid progress of society, industrial production demands have become increasingly diversified and personalized, with drastic and rapid fluctuations in demand. These new operational demands require operating systems to respond quickly, especially in the manufacturing sector.

[0003] Increasingly diverse and personalized demands have made the allocation of personnel and equipment within control systems increasingly complex. Typically, a work unit comprises several personnel and equipment, and shared needs for the same personnel or equipment often arise between different work units, resulting in resource conflicts. Multiple work units with resource conflicts cannot operate in parallel. Effectively handling resource conflicts between work units is crucial for improving order processing efficiency and is a key and pressing technical issue in this field. The core issue lies in how to accurately and rationally plan work units with resource conflicts within a limited work area to efficiently fulfill arriving order demands.

[0004] To better handle resource conflicts between job units, the concept of families from batch scheduling has been introduced to describe the complexity of conflicts. Job units with resource conflicts are assigned to different families; job units in different families cannot run simultaneously; job units within the same family do not have resource conflicts and can run in parallel. The more families there are, the more complex the conflicts. The batch scheduling problem originated in semiconductor manufacturing and has wide applications in steel smelting, logistics, and other fields. In related research, Chen et al. proposed an online batch scheduling method with a contention ratio of 10 / 3 in their paper "On-Line scheduling a batch processing system to minimize total weighted job completion time," aiming to minimize the sum of weighted completion times. Nong et al., in their paper "The single-machine parallel-batching on-linescheduling problem with family jobs to minimize makespan," proposed a batch scheduling method with a contention ratio of 2, aiming to minimize the maximum completion time and considering that jobs belonging to different families will arrive at two fixed time points. Fu et al., in their paper "On-line scheduling on an unbounded parallel batchmachine to minimize makespan of two families of jobs," proposed a competition ratio of [missing information - likely a specific ratio or parameter]. The batch processing scheduling method.

[0005] Furthermore, the rapidly changing and volatile operational demands significantly shorten order delivery cycles, placing high demands on the rapid response capabilities of operational systems. Traditional operational system research typically considers static models and employs static programming methods. Before the operational cycle begins, all information regarding the operational demands is known, including the personnel composition and internal division of labor within the operational unit, as well as the construction time and sequence. In recent years, research on how to cope with rapidly changing and volatile operational demands has gradually increased. Faced with such demands, operational systems must possess strong responsiveness, capable of responding quickly or even instantly to newly arriving demands. Real-time decision-making models can better adapt to rapidly changing demands; that is, making dynamic decisions over time to provide a rapid response to volatile demands.

[0006] The demand for rapid response from volatile and rapidly changing demands presents a significant challenge. Ideally, each incoming order should trigger the immediate creation of a corresponding work unit. However, in actual production, this is often difficult to achieve. For example, if there are no available workspaces in the current work area for a new work unit, or if there are resource conflicts between waiting and already created work units, the waiting work unit may have to wait for a period before completion. Since waiting is unavoidable, deciding when to wait and the length of that wait becomes crucial for improving the overall efficiency of the operational system and enabling rapid response to fluctuating demands. Summary of the Invention

[0007] The purpose of this invention is to address the shortcomings and deficiencies of existing technologies. To solve the technical problems of traditional operating systems' inability to quickly respond to drastically fluctuating and rapidly changing order demands, and how to achieve rapid, accurate, and rational planning and processing of work units with resource conflicts under diverse and personalized demands, this invention proposes a real-time batch processing system and method for weighted orders with resource conflicts. This invention effectively addresses diverse, personalized, drastically fluctuating, and rapidly changing demands, enabling rapid processing of weighted orders and improving the rapid response capability of the control operating system.

[0008] To solve the above-mentioned technical problems, the present invention adopts the following technical solution.

[0009] On one hand, this invention proposes a real-time batch processing system for weighted orders to address resource conflicts. This system includes an order receiving module, a work area monitoring module, an order information processing module, a job unit selection module, and a job unit construction module. For example... Figure 1 As shown.

[0010] The order receiving module is used to update the order information arriving in the system in real time.

[0011] The work area monitoring module is used to update the status of work units within the work area in real time;

[0012] The order information processing module is used to generate job units to be built based on the order information to be processed;

[0013] The job cell selection module is used to select the job cells to be built in the next batch;

[0014] The job cell construction module is used to determine whether to immediately construct the selected job cells and to make decisions on the construction order and construction time of the job cells.

[0015] This invention takes into account a series of weighted orders arriving over time, and unfolds the work in the form of units, with one order corresponding to one work unit. The relevant concepts are explained as follows:

[0016] For work units whose worker composition and internal division of labor have been determined according to the order task, they are divided into work units that have been built and work units that are yet to be built, depending on whether they have been built and started in the work area.

[0017] The time required to determine the composition of the workforce and the internal division of labor is called the determination time; the time required for a work unit to process the corresponding order task is called the processing time; the time required to construct a work unit is called the construction time; and the time required for a work unit to complete the corresponding order task is called the completion time.

[0018] When a work unit needs the same personnel or equipment, resource conflicts often occur between work units. Based on the conflict relationship, the work units are divided into different families. There are no conflicts between work units in the same family, and work units in different families cannot work in parallel. The more complex the conflict, the more families are divided.

[0019] The working principle of this system is as follows: when an idle job position appears, with the goal of minimizing the sum of weighted completion times, the job units to be built are divided into different batches according to their families. Job units in the same batch belong to the same family. The next batch of units to be worked on is quickly selected from these batches, and decisions are made on the time and order of the job unit construction.

[0020] The system of the present invention has the following features:

[0021] (1) Orders arrive over time, and their relevant information is unknown before arrival;

[0022] (2) There is a one-to-one correspondence between orders and work units. One order is completed by one work unit. One work unit is only responsible for processing one order. The work units to be built are updated in real time as the orders arrive. Only after the personnel composition and internal division of labor of the corresponding work unit are determined when the order arrives can the corresponding work unit be built.

[0023] (3) Orders have different weights, and the work units to be built are assigned corresponding weights according to the corresponding order tasks;

[0024] (4) There are resource conflicts between the job units to be built. Job units of different families cannot carry out parallel operations in the same time period. Job units that carry out parallel operations in the same time period belong to the same family.

[0025] (5) The number of locations available for building work units within the work area is limited, and only one work unit can be accommodated at a location. At the same time, the number of work units built does not exceed the maximum number of work units that can be accommodated within the work area.

[0026] (6) The work unit is non-interruptible and non-preemptive. Once the construction is completed, its work process cannot be interrupted until the corresponding order task is completed before the work can be stopped.

[0027] (7) Uncontrollable factors that may cause work stoppages, such as insufficient materials or equipment failures, are not taken into account.

[0028] In this system, the order receiving module updates the order information in real time as it arrives over time. Before an order arrives, its relevant information is unknown. After an order arrives, its corresponding weight, arrival time, required product type and quantity are entered into the order receiving module.

[0029] The work area monitoring module monitors the work units within the work area in real time and determines whether there are any vacant work locations within the work area where work units can be built. Whenever a new work unit is built, its relevant information is entered into the work area monitoring module, including the corresponding order information, the information of the workers and equipment involved, as well as the determination time and processing time of the work unit. Whenever a work unit finishes its work and completes the delivery of the corresponding order task, the work area monitoring module ends the monitoring of that work unit, its corresponding work location is released, and the status is updated to vacant.

[0030] In the order information processing module, work units to be built are generated based on the order information to be processed. The weight of the work unit is equal to the weight of the corresponding order. When the work units to be built have common needs for the same operator or equipment, a resource conflict occurs. Based on the conflict relationship, the work units to be built are divided. Work units with resource conflicts cannot work in parallel and will be divided into different families. There are no conflicts between work units in the same family, and they can work in parallel.

[0031] In the job unit selection module, resource conflicts between job units are considered when selecting job units for the next batch of construction. The job units to be constructed are divided according to their families, resulting in several batches to be constructed. If there are already constructed job units in the work area, the batch to be constructed that belongs to the same family as the already constructed job units is selected. If the work area is empty, the batch with the smallest weighted processing time is selected.

[0032] In the job unit construction module, the system combines information about the job unit to be constructed and the job units already constructed in the work area to determine whether to immediately construct the job units in the selected batch to be constructed. If the determination passes, the job units in the batch to be constructed are constructed in descending order of processing time to determine their construction time and thus the completion time. If the determination fails, the job units in the batch to be constructed are not constructed.

[0033] Beneficial effects

[0034] Compared with existing technologies, this invention addresses the resource conflict problem between work units under diverse and personalized order demands. It considers the requirement for rapid response capability of the work system due to drastic fluctuations and rapid changes in demand. Through the mutual coordination between the order receiving module, work area monitoring module, order information processing module, work unit selection module, and work unit construction module, it can quickly, accurately, and reasonably plan work units with resource conflicts in real time within a limited work area. This effectively improves the work efficiency of the work system, shortens the work cycle of the work system, and realizes rapid response to fluctuating order demands. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the overall structure of the system described in this invention;

[0036] Figure 2 This is a flowchart illustrating the implementation method of the system of the present invention;

[0037] Figure 3 The figure shows a comparison of simulation results between existing methods and the present invention. Detailed Implementation

[0038] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0039] This invention considers weighted orders and deploys operations in the form of units. To address resource conflicts between work units, and with the goal of minimizing the sum of weighted completion times, it proposes a strategy for real-time batch processing of weighted orders. This strategy enables rapid, reasonable, and accurate planning of work units with resource conflicts within a limited work area.

[0040] The indexes and variable definitions involved in the embodiments are shown in Table 1:

[0041] Table 1 Index and Variable Definitions

[0042]

[0043]

[0044] The target quantity is described as follows:

[0045] min∑ j w j c j

[0046] Among them, w j This indicates the weight of the work unit, which is equal to the weight of the corresponding order; c j The time it takes for a work unit to complete its corresponding task is called the completion time. j is the order index, i.e., the work unit index, j = 1, 2, ..., n.

[0047] like Figure 1 , Figure 2 As shown, a weighted order real-time batch processing system for resource conflicts includes the following steps in its operation:

[0048] Step S100: The order receiving module updates the information waiting to be processed at time t, including its corresponding weight, arrival time, required product type and quantity;

[0049] Step S200: The work area monitoring module monitors the work area in real time and updates the status of the work units in the area at time t. If there are work positions that are idle and there are several work units with a determined personnel composition and internal division of labor but which have not yet been constructed, proceed to step S300; otherwise, t = t + 1, proceed to step S100.

[0050] Step S300: The order information processing module generates a work unit to be built based on the order information to be processed, determines the personnel composition and specific division of labor within the work unit, including obtaining the determined time r of the work unit. j Processing time, weight w j ;

[0051] Within the job unit selection module, the processing method is as follows:

[0052] Step S401: Based on the conflict relationship, divide the work units to be built into several non-overlapping batches to be built. Work units in the same batch belong to the same family, and work units in different batches belong to different families.

[0053] Step S402: If a job cell has been built, proceed to step S403; if no job cell has been built, i.e. the job area is empty, proceed to step S404.

[0054] Step S403: If there is a batch to be built that belongs to the same family as the already built job unit, record the batch to be built as B. * Proceed to step S404; if there is no batch to be built that belongs to the same family as the already built job unit, t = t + 1, proceed to step S100;

[0055] Step S404: Calculate the weighted sum of processing times ∑w for each batch to be built, including all job units. j p j p j The time required for a work unit to process the corresponding order task is called the processing time; the batch to be built with the smallest weighted sum of processing times is selected and denoted as B. * ;

[0056] Specifically, it includes the following steps:

[0057] Step S501: Calculate B * The total task time p(B) is constructed by organizing the internal task units in descending order of processing time. * ), that is, B * The difference between the maximum completion time and the minimum construction time of an internal work unit, if B * satisfy:

[0058]

[0059] Proceed to step S502; otherwise, t = t + 1, proceed to step S100.

[0060] Step S502: Place B * The job cells are constructed in descending order of processing time. Output the weighted sum of the completion times of all job cells constructed so far, ∑. j w j c j , t = t + 1, proceed to step S100.

[0061] This invention can effectively improve the working efficiency of the operating system, shorten the operating cycle of the operating system, and achieve rapid response to fluctuating demands.

[0062] Experimental verification

[0063] To more clearly and explicitly describe the beneficial effects of the present invention compared to the prior art, a simulation experiment is conducted below using specific data and following the execution steps in the embodiments to demonstrate the weighted order real-time batch processing system and method for resource conflicts proposed in this invention. Experimental results show that the present invention has good adaptability to diverse, personalized, volatile, and rapidly changing operational demands, and can quickly and effectively handle resource conflicts between operational units, achieving rapid response to operational needs.

[0064] The experimental case parameter settings for simulation in this invention are as follows: there are n orders to be processed, each order corresponds to a work unit, that is, there are n work units, denoted as J1~Jn; there are m locations in the work area that can be used to place work units, denoted as L1~Lm; the number of families obtained by dividing work units according to conflict relationships is f, denoted as F1~Ff.

[0065] The parameter settings for the work unit are as follows:

[0066] The timeframe for determining the personnel composition and specific internal division of labor within a work unit. j It is generated by a Poisson distribution with parameter λ, where λ represents the average number of job units to be built per unit time, that is, the average number of orders arriving per unit time. The larger λ is, the higher the frequency of order arrival.

[0067] The processing time p required for a work unit to complete the corresponding order task j It is generated by a Gaussian distribution with parameters μ and σ, where μ represents the average processing time of the unit. With μ constant, the larger the parameter σ, the greater the fluctuation in order size.

[0068] Since the focus of this invention is on how to perform real-time batch processing of weighted orders within a limited working area, and how to quickly and accurately plan work units with resource conflicts in order to minimize the sum of weighted completion times, the simulation experiment does not consider how specific personnel allocation and work division are completed within the work unit. Each work unit is randomly divided into f families, and the weight of each work unit is randomly generated from {1,2,3}.

[0069] The simulation environment is as follows:

[0070] (1) Operating System: Windows 10 Professional Edition

[0071] (2)CPU: 11th Gen Intel(R)Core(TM)i9-11900K@3.50GHz

[0072] (3) Memory: 96GB

[0073] (4) Development environment: Visual Studio Code 1.53.0

[0074] (5) Programming language: Python 3.7.0

[0075] Given a parameter combination of n=15, m=5, λ=0.3, cv=σ / μ=0.5, and F=4, provide a specific example with 15 orders. In this example, the determination time r of each unit is... j Processing time pj Weight w j The clan to which it belongs, and the construction time s j The construction locations are shown in Table 2; the start time and family to which each batch of construction work units belong are shown in Table 3. Figure 1 The steps shown are used to calculate the weighted sum of completion times ∑ in this embodiment. j w j c j =1206.

[0076] Table 2 shows the determination time, processing time, weight, family, construction time, and location of each unit.

[0077]

[0078]

[0079] Table 3. Each batch of units, their start construction time, and the families they belong to.

[0080]

[0081] No research results have been published regarding the weighted order real-time processing problem considering resource conflicts discussed in this invention. In a large-scale comparative experiment, referring to existing research, an optimal online scheduling method for a single batch processor capable of simultaneously processing a finite number of jobs is identified, aiming to minimize the sum of weighted completion times. By replacing the batch division in this method with the family division in this invention, an improved method applicable to the problem studied in this invention is obtained. This improved method is compared with this invention to explore its performance under diverse, personalized, volatile, and rapidly changing job requirements.

[0082] Based on the parameter values ​​given in Table 4, there are 528 different parameter combinations (m, λ, σ, f). For each parameter combination, 1000 test cases are generated, totaling 528,000 test cases. Under each parameter combination, the average ratio of the sum of the weighted completion times obtained by the improved method and the present invention is calculated. When the ratio is greater than 1, it can be concluded that the present invention is superior; the larger the ratio, the better the performance of the present invention.

[0083] Table 4. Simulation Experiment Parameter Values

[0084]

[0085]

[0086] Experimental results show that in 86.57% of the test cases, the present invention is superior to the improved method; in the remaining test cases, the difference between the present invention and the improved method is less than 6.33%.

[0087] Figure 3 This paper presents a comparison of the weighted average completion time ratios between existing methods and the present invention. The X-axis represents (λ, σ), sorted in ascending order with λ as the primary key and σ as the secondary key; the Y-axis represents the average ratio of the weighted average completion time obtained by the improved method and the present invention. Figure 2 As can be seen, the following conclusions can be drawn in different working areas: as the parameter λ increases, that is, as the order arrival frequency increases, the performance advantage of the present invention becomes more obvious; as the parameter σ increases, that is, as the order size fluctuates more drastically, the performance advantage of the present invention becomes more obvious.

[0088] In summary, this invention is highly adaptable to diverse, personalized, volatile, and rapidly changing operational needs. It can process weighted orders in real time within a limited work area by constructing operational units, and quickly and accurately plan operational units with resource conflicts.

[0089] It should be understood that this embodiment is merely a specific example of the implementation of the present invention and should not be considered a limitation on the scope of protection of the present invention. Any equivalent modifications or alterations to the above content without departing from the spirit and scope of the present invention should be included within the scope of protection claimed by the present invention.

Claims

1. A real-time batch processing method for a weighted order real-time batch processing system with resource conflicts, characterized in that, It includes an order receiving module, a work area monitoring module, an order information processing module, a job unit selection module, and a job unit construction module; The order receiving module is used to update the order information arriving in the system in real time; The work area monitoring module is used to update the status of work units within the work area in real time; The order information processing module is used to generate job units to be built based on the order information to be processed; The job cell selection module is used to select the job cells to be built in the next batch; The job cell construction module is used to determine whether to immediately construct the selected job cells and to make decisions on the construction order and construction time of the job cells. The target quantity is described as follows: in, The weight of the work unit is equal to the weight of the corresponding order; The time it takes for a work unit to complete its corresponding task is called the completion time. This is the order index, i.e., the job cell index. ; Step S100: Order receiving module update Information that is always waiting to be processed includes its corresponding weight, arrival time, required product type and quantity; Step S200: The work area monitoring module monitors the work area in real time and updates... Within the specified time frame, if there are idle work locations and several work units with defined personnel and internal roles but not yet completed, proceed to step S300; otherwise, ... Proceed to step S100; Step S300: The order information processing module generates a work unit to be built based on the order information to be processed, determines the personnel composition and specific division of labor within the work unit, including obtaining the determination time of the work unit. Processing time, weight ; In the job unit selection module, the processing method is as follows: Step S401: Based on the conflict relationship, divide the work units to be built into several non-overlapping batches to be built. Work units in the same batch belong to the same family, and work units in different batches belong to different families. Step S402: If a job cell has been built, proceed to step S403; if no job cell has been built, i.e. the job area is empty, proceed to step S404. Step S403: If there is a batch to be built that belongs to the same family as the already built job unit, record the batch to be built as... Proceed to step S404; if no batch to be built exists that belongs to the same family as the already built job cell, Proceed to step S100; Step S404: Calculate the weighted sum of processing times for each job unit contained in each batch to be built. , The time required for a work unit to process the corresponding order task is called the processing time; the batch to be built with the smallest weighted sum of processing times is denoted as... ; Step S501: Calculation The total task time is constructed by organizing the internal task units in descending order of processing time. ,Right now The difference between the maximum completion time and the minimum construction time of an internal work unit, if satisfy: Proceed to step S502, otherwise, Proceed to step S100; where The number of groups is the number of groups. For the current moment; Step S502: ... The job cells are constructed in descending order of processing time, and the weighted sum of the completion times of all job cells constructed so far is output. , Proceed to step S100.

2. The real-time batch processing method as described in claim 1, characterized in that: Orders arrive over time, and their relevant information is unknown until they arrive. There is a one-to-one correspondence between orders and work units. One order is completed by one work unit, and one work unit is only responsible for processing one order. The work units to be built are updated in real time as orders arrive. Once the order arrives and the personnel composition and internal division of labor of the corresponding work unit are determined, the corresponding work unit can be built. Orders have different weights, and the job units to be built are assigned corresponding weights based on the order tasks they correspond to. There are resource conflicts between the job units to be built. Job units of different families cannot carry out parallel operations at the same time period. Job units that carry out parallel operations at the same time period belong to the same family. The number of locations available for building work units within the work area is limited, and each location can only accommodate one work unit. At any given time, the number of work units built cannot exceed the maximum number of work units that can be accommodated within the work area. The work unit is non-interruptible and non-preemptive. Once the construction is completed, its work process cannot be interrupted until the corresponding order task is completed before the work can stop. Uncontrollable factors that could cause work stoppages are not taken into account; For work units whose worker composition and internal division of labor have been determined according to the order task, they are divided into work units that have been built and work units that are yet to be built, depending on whether they have been built and started in the work area. The time required to determine the composition of the workforce and the internal division of labor is called the determination time; the time required for a work unit to process the corresponding order task is called the processing time; the time required to construct a work unit is called the construction time; and the time required for a work unit to complete the corresponding order task is called the completion time. When the work units to be built have common needs for the same workers or equipment, resource conflicts arise between work units. Based on the conflict relationship, the work units are divided into different families. There are no conflicts between work units in the same family, and work units in different families cannot work in parallel. The more complex the conflict, the more families are divided. When an idle job location is available, with the goal of minimizing the sum of weighted completion times, the job units to be built are divided into different batches according to their families. Job units in the same batch belong to the same family. The next batch of units to be worked on is quickly selected from these batches, and decisions are made on the time and order of the job unit construction.

Images