Intelligent logistics intelligent scheduling algorithm

Abstract

The application discloses an intelligent logistics intelligent scheduling algorithm, which is applied to the logistics field or the information processing field; and solves the technical problem of low logistics scheduling efficiency; and adopts the technical scheme of an intelligent logistics intelligent scheduling algorithm, which comprises the following steps: (S1) constructing an intelligent logistics system, and calculating the minimum cost and time of reverse logistics of materials; (S2) adopting an ABC algorithm to globally search materials and corresponding sources, and scheduling and planning a shortest logistics path; and (S3) adopting an AQM algorithm to improve the ABC algorithm, calculating logistics data changes in the whole intelligent logistics system, and optimizing the ABC algorithm; the ABC algorithm can perform logistics scheduling work at the minimum cost, the AQM algorithm is used to eliminate the influence of data fluctuation generated by material management, the material management is restored to a queue type management mode, and therefore, the purpose of stable optimization scheduling of the ABC algorithm is achieved, and the efficiency of logistics scheduling is greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of logistics or information processing, and more specifically to an intelligent logistics intelligent scheduling algorithm; Background Art

[0002] In today's high-tech society, facing the depletion of natural energy and resources, as well as the ecological pollution caused by the disposal of waste and obsolete products, innovative reverse logistics capabilities play a significant role in empowering entities to better utilize ecological energy and resources and achieving a clear balance between the financial system and the ecological spectrum. Online shopping, mail order, and after-sales service contribute significantly to the returns of these organizations, yet these organizations have failed to further focus on improving these returns. Given the varying legislative measures aimed at protecting ecological and service benefits, most organizations have adopted a reverse flow approach, that is, flow from customers to recycling centers located within their logistics networks. Consumer product recycling is no longer a novelty; waste paper recycling, soft drink bottle collection, and metal scrap brokerage have become commonplace in everyday life. Reverse logistics, in turn, is driven by a variety of factors, including competitive and marketing objectives, direct financial considerations, and growing concerns about the ecosystem. Importantly, it has both an ecological and value-added dimension, thus reducing the time required for reverse logistics processes through the use of optimized scheduling algorithms.

[0003] A prior art heuristic algorithm constructs an integer programming model that incorporates the different cost components incurred during the disassembly process to minimize the costs associated with the disassembly procedure. While this method can produce an ideal schedule, it is inadequate for practical estimation problems due to unreasonable processing time. Another invention proposes an imaginative mixed-integer direct programming strategy that is promising for configuring and scheduling supply chains with reverse flows. Despite simultaneously evaluating generation, transmission, and reverse logistics capabilities, errors often occur during logistics scheduling. Summary of the Invention

[0004] In response to the above problems, the present invention discloses an intelligent logistics scheduling algorithm that can deploy network servers and realize the analysis and processing of network data.

[0005] In order to achieve the above technical effects, the present invention adopts the following technical solutions:

[0006] An intelligent logistics intelligent scheduling algorithm, the steps include:

[0007] (S1) Build an intelligent logistics system to calculate and minimize the cost and time of reverse logistics of materials;

[0008] (S2) Using the ABC algorithm to globally search for materials and their corresponding sources, scheduling and planning the shortest logistics path;

[0009] (S3) The AQM algorithm is used to improve the ABC algorithm, calculate the changes in logistics data in the entire intelligent logistics system, and optimize the settings of the ABC algorithm.

[0010] As a further technical solution of the present invention, the calculation method for minimizing the cost and time of reverse logistics of component materials is shown in formula (1):

[0011]

[0012] In formula (1), T1 represents the time required to disassemble the i-th destructive component material, T2 represents the time required to disassemble the i-th non-destructive component material; t2 represents the average time used to disassemble the i-th destructive component material; t2 represents the average time used to disassemble the i-th non-destructive component material, E represents the total re-manufactured scrapped products, and D is the non-destructive disassembly efficiency; the main goal of component materials is to minimize the total disassembly time and total cost. The function of the total material disassembly time is shown in equation (2):

[0013]

[0014] In formula (2), T3 represents the total time required for material disassembly; the function of the total material cost is shown in equation (3):

[0015]

[0016] In formula (3), M represents the function of total material cost, A i Table 1 shows the unit recovery cost of material i, B i represents the unit purchase cost of material i, C i represents the unit disposal cost of material i, R represents the total quantity of purchased materials, and U represents the total quantity of disposed materials.

[0017] As a further technical solution of the present invention, the ABC algorithm steps include:

[0018] (S21) In the initial stage of the ABC algorithm, the material source data is randomly generated, F = {F1, F2, ..., F i}, F represents the material source data, i represents the serial number;

[0019] (S22) In the global search phase of the ABC algorithm, each logistics scheduling scheme is randomly changed to obtain a new logistics scheduling scheme, which will perform neighborhood search on the material data. The new logistics scheduling scheme is obtained by using the differential expression from F i Created as shown in formula (4):

[0020] s i =F i+r×(F i -F k ) (4)

[0021] In formula (4), s i represents a new logistics scheduling plan, k belongs to {1,2,…,i} and k is not equal to i, r represents a uniform random number in the range [-1,1]; once s is obtained i , will be s i To evaluate and compare: If s i The fitness is better than the old logistics scheduling solution x i , the employed bees will forget the old logistics scheduling plan and remember the new logistics scheduling plan; otherwise, the next global search will be carried out;

[0022] (S23) In the interactive phase of the ABC algorithm, after the global domain search is completed, the logistics scheduling plan shares the logistics scheduling plan with the adjacent material source data, and then each material source data will select the logistics scheduling plan in a probabilistic manner, and the material source data picks up the old logistics scheduling plan x i The probability P i As shown in formula (5):

[0023]

[0024] In formula (5), fi is the fitness of the old logistics scheduling plan xi; once the material source data selects the food source x i , the material source data will also be used for x i Perform local search, the same as step (S2), if the modified logistics scheduling scheme has better fitness, it will replace x i ;

[0025] (S24) In the logistics scheduling plan generation stage of the ABC algorithm, if the quality of the logistics scheduling plan cannot be improved after the global search of the ABC algorithm, it is assumed that the logistics scheduling plan is abandoned, and the logistics scheduling plan will randomly generate material source data and return to step (S2) until the number of iterations.

[0026] As a further technical solution of the present invention, the AQM algorithm improves the ABC algorithm and calculates the performance index of the entire intelligent logistics system according to the changes in logistics data during the operation of the intelligent logistics system as shown in formula (6):

[0027]

[0028] In formula (6), J represents the operating performance of the intelligent logistics system, Ω represents the trajectory function of the material operation change, Y represents the prediction of the material data change law, and U represents the optimal logistics data change performance. According to the amount of material management incoming data, the logistics queue optimization amount is derived as shown in formula (7):

[0029] U=(U1QB+S) -1 BQ[Ω-Γ] (7)

[0030] In formula (7), U1 represents the optimized amount of queues within the logistics time, Q represents the difference in logistics scheduling after optimization, B represents the amount of materials required to maintain the stable compensation of the intelligent logistics system, S represents the change margin of the intelligent logistics system, and Г represents the total amount of data that needs to be converted during the operation of the intelligent logistics system. By adjusting the trajectory function of material management changes, the ABC algorithm scheduling is made more reasonable. The improved change trajectory is expressed as:

[0031]

[0032] In formula (8), q represents the minimum change in logistics data, β represents the change law function of logistics data, R0 represents the total amount of logistics data, N represents the total number of logistics data transport batches, and H represents the optimized logistics data division. According to the improved logistics data change trajectory, the actual amount of logistics data is substituted into the calculation, and the change law function is obtained as follows:

[0033] β(k)=β(k-1)+μq(k) (9)

[0034] In formula (9), k represents the logistics data transportation batch, and μ represents the change in the logistics data weight. The AQM algorithm is used to eliminate the fluctuation effect of material management, so that material management can be restored to a queue management mode, thereby achieving the purpose of ABC algorithm optimization scheduling stability.

[0035] As a further technical solution of the present invention, the intelligent logistics system includes:

[0036] The procurement module is used for enterprise configuration of shipping locations for suppliers and third-party personnel to execute material procurement;

[0037] Inbound dispatch system for intelligent logistics scheduling and inbound call information evaluation;

[0038] Inbound call history database, used to store inbound call history data;

[0039] Account information database, used to store account information of purchasing users, suppliers and third-party personnel;

[0040] Inbound call receiving equipment, used to connect to external communication signals through optical fiber interfaces, send data queries to pass caller information to the inbound dispatch system, and assign inbound calls to answering queues;

[0041] Answer queue, used to distribute incoming call signals to corresponding user call devices;

[0042] Among them, the procurement module is unidirectionally connected to the inbound scheduling system, the inbound scheduling system, the inbound call history database and the account information database are bidirectionally connected, the optical fiber interface is unidirectionally connected to the inbound telephone receiving device, the inbound telephone receiving device is bidirectionally connected to the inbound evaluation system, the inbound telephone receiving device is unidirectionally connected to the answering queue, and the answering queue is unidirectionally connected to the user call device.

[0043] As a further technical solution of the present invention, the procurement module includes:

[0044] The Evaluation Module, the primary analytical tool used to evaluate vendor proposals and determine contract awards;

[0045] Summary module, used to convey the complete contract to the purchasing client and third-party intermediaries for review;

[0046] Supplier database, which stores descriptive information about logistics providers, shipper requirements, and approved suppliers;

[0047] Proposal database, which stores descriptive information on business activities and logistics contracts;

[0048] Among them, the summary module is bidirectionally connected to the procurement user, the third-party personnel are unidirectionally connected to the supplier database, the evaluation module and the summary module, the supplier database is unidirectionally connected to the proposal database, the supplier database is bidirectionally connected to the logistics supplier, the proposal database is unidirectionally connected to the summary module, the proposal database is bidirectionally connected to the evaluation module, and the proposal database is bidirectionally connected to the logistics supplier.

[0049] As a further technical solution of the present invention, the inbound scheduling system includes a scheduling module, a call evaluation module, a modeling module and an intelligent scheduling algorithm.

[0050] As a further technical solution of the present invention, the account information includes co-ownership information, financial strength information, security review information, operational capability information, business information, operational contact information, real ownership information and debt capacity information.

[0051] The beneficial positive effects of the present invention are:

[0052] Different from conventional technologies, the present invention can provide an efficient, responsive and proactive information system arrangement. Through the ABC algorithm, logistics scheduling can be carried out at the lowest cost. The AQM algorithm is used to eliminate the impact of fluctuations caused by material management, so that material management can be restored to a queue-type management mode, thereby achieving the purpose of optimizing scheduling stability through the ABC algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive work, among which:

[0054] Figure 1 A diagram of an intelligent logistics scheduling algorithm is presented;

[0055] Figure 2 The structural diagram of the intelligent logistics system is shown;

[0056] Figure 3 The internal structure diagram of the procurement module is shown;

[0057] Figure 4 A flow chart showing the scheduling of inbound calls to be responded to by agents;

[0058] Figure 5 A comparison graph of the processing efficiency of the three algorithms is shown. DETAILED DESCRIPTION

[0059] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the embodiments described herein are only used to illustrate and explain the present invention and are not intended to limit the present invention.

[0060] like Figure 1 As shown, an intelligent logistics intelligent scheduling algorithm includes the following steps:

[0061] (S1) Build an intelligent logistics system to calculate and minimize the cost and time of reverse logistics of materials;

[0062] In particular, the creation of sub-devices is steadily increasing, leading to significant electronic waste. This is a result of the current demand for new innovations within a short timeframe. This situation forces manufacturers to create new items within a very short timeframe, necessitating the disposal or disassembly of existing items. In this context, manufacturers are looking to remanufacture (recycle) scrapped products to meet component needs for new product production. In this regard, in our previous work, we demonstrated a method for selecting the optimal amount of recycled products, thereby reducing the overall cost of reverse logistics.

[0063] (S2) Using the ABC algorithm to globally search for materials and their corresponding sources, scheduling and planning the shortest logistics path;

[0064] In a specific embodiment, this paper proposes a novel reverse logistics method for remanufacturing scrapped items, based on an optimized scheduling algorithm. Previous work has identified the optimal quantity of recycled remanufactured scrapped products for order-based disassembly. This paper demonstrates a procedure for rationally scheduling machine disassembly products, thereby reducing the total time and cost required for reverse logistics operations. The proposed architecture utilizes the Artificial Bee Colony (ABC) algorithm. Compared to existing scheduling algorithms, the proposed framework achieves superior performance.

[0065] (S3) The Active Queue Management (AQM) algorithm is used to improve the ABC algorithm, calculate the changes in logistics data in the entire intelligent logistics system, and optimize the ABC algorithm.

[0066] In a specific embodiment, order disassembly is a key process in reverse logistics for remanufactured scrap products. A proper order disassembly process can reduce the overall cost of reverse logistics. Predicting the exact yield of components from remanufactured scrap is difficult. On the other hand, reverse logistics can maximize the yield from remanufactured scrap. Furthermore, to meet demand, parts must be purchased from external suppliers. If actual yields are lower than expected, parts must be purchased to meet demand. Alternatively, if yields are higher than expected, excess components must be disposed of. Consequently, a minimal or extreme forecast can lead to misfortune in the remanufactured scrap reverse logistics operation. Consequently, the forecast should approximate the actual yield.

[0067] In a specific embodiment, the calculation to minimize the cost and time of reverse logistics of component materials is shown in formula (1):

[0068]

[0069] In formula (1), T1 represents the time required to disassemble the i-th destructive component material, T2 represents the time required to disassemble the i-th non-destructive component material; t2 represents the average time used to disassemble the i-th destructive component material; t2 represents the average time used to disassemble the i-th non-destructive component material, E represents the total re-manufactured scrapped products, and D is the non-destructive disassembly efficiency; the main goal of component materials is to minimize the total disassembly time and total cost. The function of the total material disassembly time is shown in equation (2):

[0070]

[0071] In formula (2), T3 represents the total time required for material disassembly; the function of the total material cost is shown in equation (3):

[0072]

[0073] In formula (3), M represents the function of total material cost, A i Table 1 shows the unit recovery cost of material i, B i represents the unit purchase cost of material i, C i represents the unit disposal cost of material i, R represents the total number of purchased materials, and U represents the total number of disposed materials. Equation (3) is the total time required for the order disassembly process in the remanufacturing scrap reverse logistics. Then, the goal of this work is to minimize the total time and total cost required for the disassembly process. Minimizing the total time function is a complex task in general because it may contain multiple products and components. Therefore, optimization is the preferred method to solve the scheduling task. In this work, we choose the artificial bee colony (ABC) algorithm optimization to solve the scheduling problem.

[0074] The ABC algorithm consists of three types of artificial bees: employing bees, hired bees, spectators, and scouts. Employed bees collect food sources from a global search. Employed bees provide data to spectator bees, who wait in the hive and watch the employed bees' dance. The spectator bees then select a food source that likely corresponds to the properties of the food source. In this way, good food sources attract more bees than bad ones. Scouting bees arbitrarily scan the hive area for new food sources. When a scout or spectator bee discovers a food source, it is hired. When a food source is completely abused, all associated employed bees abandon the position and may become scouts again. Thus, scout bees engage in the "investigation" profession, while employing and spectator bees engage in the "abuse" profession. This further encourages the scout bee stage to work. The steps of the ABC algorithm include:

[0075] (S21) In the initial stage of the ABC algorithm, the material source data is randomly generated, F = {F1, F2, ..., F i}, F represents the material source data, i represents the serial number;

[0076] (S22) In the global search phase of the ABC algorithm, each logistics scheduling scheme is randomly changed to obtain a new logistics scheduling scheme, which will perform neighborhood search on the material data. The new logistics scheduling scheme is obtained by using the differential expression from F i Created as shown in formula (4):

[0077] s i =Fi +r×(F i -F k ) (4)

[0078] In formula (4), s i represents a new logistics scheduling plan, k belongs to {1,2,…,i} and k is not equal to i, r represents a uniform random number in the range [-1,1]; once s is obtained i , will be s i To evaluate and compare: If s i The fitness is better than the old logistics scheduling solution x i , employed bees will forget the old logistics schedule and remember the new one; otherwise, another global search will be conducted. In the proposed algorithm, food sources are associated with conceivable solutions to the optimization problem, and the nectar measurements of the food sources are compared with the fitness of the associated solutions. In ABC, the first half of the bee colony consists of employed bees, and the other half consists of spectators. Since only one bee is expected to be present at each food source, the number of employed bees is equal to the number of food sources. Therefore, the number of spectator bees is also equal to the number of solutions being considered. The ABC algorithm first collects randomly generated food sources. The main method of ABC can be described as follows.

[0079] (S23) In the interactive phase of the ABC algorithm, after the global domain search is completed, the logistics scheduling plan shares the logistics scheduling plan with the adjacent material source data, and then each material source data will select the logistics scheduling plan in a probabilistic manner, and the material source data picks up the old logistics scheduling plan x i The probability P i As shown in formula (5):

[0080]

[0081] In formula (5), fi is the fitness of the old logistics scheduling plan xi; once the material source data selects the food source x i , the material source data will also be used for x i Perform local search, the same as step (S2), if the modified logistics scheduling scheme has better fitness, it will replace x i ;

[0082] (S24) In the logistics scheduling plan generation stage of the ABC algorithm, if the quality of the logistics scheduling plan cannot be improved after the global search of the ABC algorithm, it is assumed that the logistics scheduling plan is abandoned, and the logistics scheduling plan will randomly generate material source data and return to step (S2) until the number of iterations.

[0083] In a specific embodiment, the AQM algorithm improves the ABC algorithm and calculates the performance index of the entire intelligent logistics system according to the changes in logistics data during the operation of the intelligent logistics system as shown in formula (6):

[0084]

[0085] In formula (6), J represents the operating performance of the intelligent logistics system, Ω represents the trajectory function of the material operation change, Y represents the prediction of the material data change law, and U represents the optimal logistics data change performance. According to the amount of material management incoming data, the logistics queue optimization amount is derived as shown in formula (7):

[0086] U=(U1QB+S) -1 BQ[Ω-Γ] (7)

[0087] In formula (7), U1 represents the optimized amount of queues within the logistics time, Q represents the difference in logistics scheduling after optimization, B represents the amount of materials required to maintain the stable compensation of the intelligent logistics system, S represents the change margin of the intelligent logistics system, and Г represents the total amount of data that needs to be converted during the operation of the intelligent logistics system. By adjusting the trajectory function of material management changes, the ABC algorithm scheduling is made more reasonable. The improved change trajectory is expressed as:

[0088]

[0089] In formula (8), q represents the minimum change in logistics data, β represents the change law function of logistics data, R0 represents the total amount of logistics data, N represents the total number of logistics data transport batches, and H represents the optimized logistics data division. According to the improved logistics data change trajectory, the actual amount of logistics data is substituted into the calculation, and the change law function is obtained as follows:

[0090] β(k)=β(k-1)+μq(k) (9)

[0091] In formula (9), k represents the logistics data transportation batch, and μ represents the change in the logistics data weight. The AQM algorithm is used to eliminate the fluctuation effect of material management, so that material management can be restored to a queue management mode, thereby achieving the purpose of ABC algorithm optimization scheduling stability.

[0092] In a specific embodiment, Figure 2As shown, the intelligent logistics system includes a procurement module, an inbound dispatch system, an inbound call history database, an account information database, an inbound call receiving device, and an answer queue. The procurement module is used by suppliers and third-party personnel to configure the shipping location for material procurement; the inbound dispatch system is used for intelligent logistics scheduling and inbound call information evaluation; the inbound call history database is used to store inbound call history data; the account information database is used to store account information for procurement users, suppliers, and third-party personnel; the inbound call receiving device is used to connect to external communication signals via an optical fiber interface, send data queries to transmit caller information to the inbound dispatch system, and assign inbound calls to the answer queue; the answer queue is used to distribute inbound call signals to corresponding user call devices. The procurement module is unidirectionally connected to the inbound dispatch system, while the inbound dispatch system, the inbound call history database, and the account information database are bidirectionally connected. The optical fiber interface is unidirectionally connected to the inbound call receiving device, which is bidirectionally connected to the inbound evaluation system. The inbound call receiving device is unidirectionally connected to the answer queue, and the answer queue is unidirectionally connected to the user call device.

[0093] In a specific embodiment, an inbound dispatch system dispatches inbound telephone calls so that agents respond in an order based in part on the predicted outcomes of the inbound telephone calls. The inbound dispatch system includes a dispatch module, a call evaluation submodule, and a modeling module, and is connected to an inbound call history database and an account information database. The modeling module constructs one or more models that use account information in the inbound call history database to predict the outcomes of inbound calls. The dispatch module applies the models to predict the outcomes of pending inbound calls and, based on the call evaluation submodule, assigns orders to agents to respond to the pending inbound calls. The modeling module 14 constructs a statistical model, and the call evaluation submodule calculates priority values ​​used by the dispatch module. The priority values ​​are based on the results of the model calculations and the results of solving an optimization problem defined by the model calculations. The inbound dispatch system and the inbound call receiving device can be integrated into a single computing platform, or can be based on a separate computing platform connected to the inbound call receiving device's proprietary application program interface, or connected to commercial application middleware. Inbound call receiving receives inbound telephone calls via a fiber optic interface. Inbound telephone call analog signals received by the inbound call receiving equipment through the fiber optic interface are distributed to the answering queue for response by the agent operating the telephone equipment.

[0094] In a specific embodiment, the inbound call receiving device transmits caller information to the inbound dispatch system through a data query or other means and awaits a response from the inbound dispatch system before assigning the inbound call to an answering queue. Furthermore, the inbound call receiving device provides the inbound dispatch system with agent activity and capacity information. For example, the inbound call receiving device can provide agent activity information. When the inbound call receiving device has sufficient capacity to respond, the inbound dispatch system immediately returns the call signal to the inbound call receiving device. In other words, if there is sufficient capacity to handle the inbound call, the inbound call is not delayed but instead routed for immediate answering. When the inbound call receiving device's capacity is insufficient, the inbound dispatch system delays the call on virtual hold until an appropriate time based on the priority value calculated by the call evaluation submodule. Regardless of whether the inbound call is delayed, the inbound dispatch system collects inbound call data and stores it in an inbound call history database. The inbound call results and predicted results are also collected and stored to provide a detailed, call-by-call history for future modeling and validation of predicted results against actual results. For example, results are generated once the inbound call is completed. For example, successful connections to agents, abandoned calls, purchases, or customer churn in billing records are all related to inbound calls.

[0095] In a specific embodiment, Figure 3 As shown, the procurement module includes an evaluation module, a summary module, a supplier database, and a proposal database. The evaluation module is the primary analytical tool for evaluating supplier proposals and determining contract awards; the summary module is used to communicate complete contracts to procurement clients and third-party intermediaries for review; the supplier database stores descriptive information about logistics suppliers, shipper requirements, and approved suppliers; and the proposal database stores descriptive information about business activities and logistics contracts. The summary module has a bidirectional connection to procurement users, while third-party personnel have a unidirectional connection to the supplier database, the evaluation module, and the summary module. The supplier database has a unidirectional connection to the proposal database, which is bidirectionally connected to logistics suppliers. The proposal database has a unidirectional connection to the summary module, and the proposal database has a bidirectional connection to the evaluation module and to logistics suppliers.

[0096] In a specific embodiment, the inbound dispatch system includes a dispatch module, a call assessment module, a modeling module, and an intelligent dispatch algorithm. The account information includes co-ownership information, financial strength information, security clearance information, operational capability information, business information, operational contact information, actual ownership information, and debt capacity information. The dispatch module holds inbound calls in a queue, which acts as a virtual hold until a response is required, and then releases the inbound call to be placed in an answer queue. Thus, the inbound dispatch system responds to inbound call receiving device queries based on the priority of the inbound call, effectively creating an ordered queue at the inbound call receiving device by delaying responses to lower-priority inbound calls. In an alternative embodiment, the dispatch module can reorder the queue directly within the inbound call receiving device to allow for real-time sorting of inbound call queues. The dispatch module obtains data for application to the caller model by performing a lookup based on caller information received from the inbound call receiving device. The caller information can include account number, zip code, area code, telephone exchange, reservation number, or other relevant information obtained from the inbound caller, such as information obtained via a VRU or by the inbound call receiving device from the inbound call, such as ANI or DNIS information. The nature of the caller information depends on the implementation of the inbound dispatch system and is usually configurable through a graphical user interface provided with the traditional receiving equipment. In addition to the caller information, the dispatch module can also query and connect data from other sources, such as Zip-4 and credit bureau sources, as well as other demographic information that can be derived from the caller information.

[0097] In a specific embodiment, the modeling module creates caller models by statistically analyzing appropriate data from an inbound call history database and an account information database. The statistical analysis performed by the modeling module builds models by correlating call outcomes (i.e., dependent variables) with information available at the time the call is received (i.e., independent variables). The end result of each model is an equation that, when evaluated, predicts the associated outcome. The application of the caller model to the caller and / or call information can be performed when the call arrives, or it can be pre-formed so that potential action codes are calculated at the beginning of the time period, allowing for a faster response when circumstances warrant.

[0098] In a specific embodiment, Figure 4The flowchart describes the process of dispatching inbound calls for agent response. This process builds a caller model from inbound call history. The inbound call history used to model the outcome of interest may be a sample of historical inbound calls of the same nature as the outcome being modeled, or it may be specially designed during the testing phase. For example, a television commercial airing in a single or limited number of television markets can represent the total target audience and be used to generate inbound calls with a volume within the call center's capacity constraints. The inbound call results from the sample commercial are then used to create a model specific to the nature of the product being sold by the advertisement. This ad-specific model is then used to schedule the time period during which the advertisement is presented to a wider audience, so that inbound calls with a greater likelihood of resulting in a purchase receive a higher priority for agent response. The inbound call is then received by a receiving device. Typically, inbound calls continue to arrive at the receiving device at a rate that varies over time. The receiving device answers the inbound call in a conventional manner and determines call and / or caller information. This information is determined by analyzing the logistics information that arrives with the inbound call.

[0099] In certain embodiments, inbound calls are next scheduled for response by an agent interfaced with the receiving device. Inbound calls with lower priority values ​​are placed on virtual hold, while outbound calls with higher priority values ​​are returned to the receiving device and placed in a queue for the agent to respond. The virtual hold length of an inbound call depends on the number of inbound calls, the capacity of the receiving device, the agent talk time per call, and the priority value of the inbound call relative to other pending inbound calls. Based on these factors, when agent resources are available and / or the maximum hold time parameter is exceeded, the inbound call is placed on virtual hold and forwarded to the receiving device in order of priority value. Alternatively, in embodiments where the receiving device can sort or change the order of the inbound queues based on available data including priority value, the receiving device's inbound queue can be reordered in real time as additional queries are received. The results of the inbound calls are then stored in an inbound call history database. The inbound history database tracks factors S, such as call success or abandonment, and the final call outcome. Call outcomes may include directly quantifiable factors S, such as purchase decisions, or less quantifiable factors S, such as account usage, cancellations, and customer satisfaction as reflected in relevant information available from the account database and other sources.

[0100] In a specific embodiment, the present invention uses Simulink software to establish a material management experimental model, and experiments are carried out under this model and data is recorded to verify the practicality of the intelligent scheduling algorithm. The computer used in the experiment of the present invention adopts a CPU memory of 64+256G and a hardware parameter of Intel core i89700KF. The model environment parameter setting data acquisition accuracy is 96%, the algorithm calculation error is not less than 2.0%, and the data communication method adopts China Unicom 5G wireless communication. This research experiment adopts Windows x86 computer system, the material management model is EPC management network, and the parameter setting meets the GEN-2 label. On this basis, experiments are carried out, and by comparing the integer programming model (Scheme 1) and the mixed integer direct programming strategy model (Scheme 2), the data results are analyzed using a computer program, and the test data table is finally displayed as shown in Table 1.

[0101] Table 1 Experimental data of three logistics models

[0102]

[0103] Analysis of the data in Table 1 reveals that the logistics model proposed in this invention achieves a maximum material data processing efficiency of 4.49 MB / s for a total material data volume of 2879 MB, and a control level of 95.4% for logistics scheduling. The integer programming model proposed in Solution 1, with a total material data volume of 2434 MB, achieves a maximum material data processing efficiency of 3.12 MB / s, and a control level of 86.2%. The mixed integer direct programming strategy model proposed in Solution 2, with a total material data volume of 2341 MB, achieves a maximum material data processing efficiency of 2.14 MB / s, and a control level of only 77.6%. This comparison reveals that the logistics model proposed in this invention is superior in all aspects.

[0104] The superiority of the intelligent scheduling algorithm of the present invention can be further verified by displaying the data processing efficiency of each logistics model. The processing efficiency of the three algorithms is simulated using MATLAB, and the efficiency comparison simulation is obtained as follows: Figure 5 As shown in the figure, a comparison shows that the intelligent scheduling algorithm of the present invention has a maximum data processing efficiency of 5.0MB / s, with a relatively fast growth rate, reaching a stable processing efficiency when the material data is 2300MB. The scheduling algorithm of solution 1 has a maximum data processing efficiency of 3.2MB / s, and the growth rate accelerates when the material data exceeds 2600MB. The scheduling algorithm of solution 2 has a maximum data processing efficiency of 2.5MB / s and a relatively slow growth rate. This analysis shows that the intelligent scheduling algorithm of the present invention has the best processing efficiency.

[0105] Although specific embodiments of the present invention are described above, those skilled in the art should understand that these specific embodiments are merely illustrative, and those skilled in the art may omit, replace, and change the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, combining the above method steps to perform substantially the same functions in substantially the same manner to achieve substantially the same results falls within the scope of the present invention. Therefore, the scope of the present invention is limited only by the appended claims.

Claims

1. An intelligent logistics scheduling algorithm, characterized by: The following steps are involved: (S1) Build an intelligent logistics system to calculate and minimize the cost and time of reverse logistics of materials; (S2) Using the ABC algorithm to globally search for materials and their corresponding sources, scheduling and planning the shortest logistics path; (S3) Using the AQM algorithm to improve the ABC algorithm, calculate the changes in logistics data in the entire intelligent logistics system, and optimize the ABC algorithm; The ABC algorithm steps include: (S21) In the initial stage of the ABC algorithm, the material source data is randomly generated, F={F1,F2,…,F i }, F represents the material source data, Indicates the serial number; (S22) In the global search phase of the ABC algorithm, each logistics scheduling scheme is randomly changed to obtain a new logistics scheduling scheme, which will perform neighborhood search on the material data. The new logistics scheduling scheme is obtained from F using the differential expression. i Created as shown in formula (4): （4） In formula (4), Represents a new logistics scheduling plan, k belongs to {1,2,…,i} and k is not equal to , Represents a uniform random number in the range [-1,1]; once obtained , will be To evaluate and compare: If The suitability is better than the old logistics scheduling solution , the employed bees will forget the old logistics scheduling plan and remember the new logistics scheduling plan; otherwise, the next global search will be carried out; (S23) In the interactive phase of the ABC algorithm, after the global domain search is completed, the logistics scheduling plan is shared with the adjacent material source data. Then each material source data will select the logistics scheduling plan in a probabilistic manner, and the material source data will pick up the old logistics scheduling plan. Probability As shown in formula (5): （5） In formula (5), It is the old logistics scheduling plan x i fitness; once the material source data selects the food source , the material source data will also be Perform local search, the same as step (S2), if the modified logistics scheduling scheme has better fitness, it will replace ; (S24) In the logistics scheduling plan generation phase of the ABC algorithm, if the quality of the logistics scheduling plan cannot be improved after the global search of the ABC algorithm, it is assumed that the logistics scheduling plan is abandoned, and the logistics scheduling plan will randomly generate material source data and return to step (S2) until the number of iterations; The AQM algorithm improves the ABC algorithm and calculates the performance index of the entire intelligent logistics system according to the changes in logistics data during the operation of the intelligent logistics system as shown in formula (6): （6） In formula (6), J represents the operating performance of the intelligent logistics system, Ω represents the trajectory function of the material operation change, Y represents the prediction of the material data change law, and U represents the optimal logistics data change performance. According to the amount of material management incoming data, the logistics queue optimization amount is derived as shown in formula (7): （7） In formula (7), U1 represents the optimized amount of queues within the logistics time, Q represents the difference in logistics scheduling after optimization, B represents the amount of materials required to maintain the stable compensation of the intelligent logistics system, S represents the change margin of the intelligent logistics system, and Г represents the total amount of data that needs to be converted during the operation of the intelligent logistics system. By adjusting the trajectory function of material management changes, the ABC algorithm scheduling is made more reasonable. The improved change trajectory is expressed as: （8） In formula (8), q represents the minimum change in logistics data, β represents the change law function of logistics data, R0 represents the total amount of logistics data, N represents the total number of logistics data transport batches, and H represents the optimized logistics data division. According to the improved logistics data change trajectory, the actual logistics data volume is substituted into the calculation, and the change law function is obtained as follows: （9） In formula (9), k represents the logistics data transportation batch, Indicates the change in the weight of logistics data; the AQM algorithm is used to eliminate the fluctuation impact of material management, so that material management can be restored to a queue management mode, thereby achieving the purpose of optimizing scheduling stability using the ABC algorithm.

2. The intelligent logistics intelligent scheduling algorithm according to claim 1 is characterized by: The calculation to minimize the cost and time of reverse logistics of materials is shown in formula (1): （1） In formula (1), T1 represents the disassembly The time required to dismantle the first destructive component material, T2 represents the time required to dismantle the first The time required to complete the delivery of a non-destructive component; Indicates the removal of The average time taken to dismantle the first destructive component; t2 represents the average time taken to dismantle the first The average time used for non-destructive component materials, E represents the total re-manufacturing of scrapped products, and D is the efficiency of non-destructive disassembly; the main goal of component materials is to minimize the total disassembly time and total cost. The function of the total material disassembly time is shown in equation (2): （2） In formula (2), T3 represents the total time required for material disassembly; the function of the total material cost is shown in equation (3): （3） In formula (3), M represents the function of total material cost, The unit recovery cost of material i is shown in the table. Indicates supplies The unit purchase cost, Indicates supplies The unit disposal cost is R, which represents the total quantity of purchased materials and U, which represents the total quantity of disposed materials.

3. The intelligent logistics scheduling algorithm according to claim 1 is characterized by: The intelligent logistics system includes: The procurement module is used for enterprise configuration of shipping locations for suppliers and third-party personnel to execute material procurement; Inbound dispatch system for intelligent logistics scheduling and inbound call information evaluation; Inbound call history database, used to store inbound call history data; Account information database, used to store account information of purchasing users, suppliers and third-party personnel; Inbound call receiving equipment, used to connect to external communication signals through optical fiber interfaces, send data queries to pass caller information to the inbound dispatch system, and assign inbound calls to answering queues; Answer queue, used to distribute incoming call signals to corresponding user call devices; Among them, the procurement module is unidirectionally connected to the inbound scheduling system, the inbound scheduling system, the inbound call history database and the account information database are bidirectionally connected, the optical fiber interface is unidirectionally connected to the inbound telephone receiving device, the inbound telephone receiving device is bidirectionally connected to the inbound evaluation system, the inbound telephone receiving device is unidirectionally connected to the answering queue, and the answering queue is unidirectionally connected to the user call device.

4. The intelligent logistics system according to claim 3, characterized in that: The procurement module includes: The Evaluation Module, the primary analytical tool used to evaluate vendor proposals and determine contract awards; Summary module, used to convey the complete contract to the purchasing client and third-party intermediaries for review; Supplier database, which stores descriptive information about logistics providers, shipper requirements, and approved suppliers; Proposal database, which stores descriptive information on business activities and logistics contracts; Among them, the summary module is bidirectionally connected to the procurement user, the third-party personnel are unidirectionally connected to the supplier database, the evaluation module and the summary module, the supplier database is unidirectionally connected to the proposal database, the supplier database is bidirectionally connected to the logistics supplier, the proposal database is unidirectionally connected to the summary module, the proposal database is bidirectionally connected to the evaluation module, and the proposal database is bidirectionally connected to the logistics supplier.

5. The intelligent logistics system according to claim 3, characterized in that: The inbound scheduling system includes a scheduling module, a call evaluation module, a modeling module and an intelligent scheduling algorithm.

6. The intelligent logistics system according to claim 3, characterized in that: The account information includes co-ownership information, financial strength information, security review information, operational capability information, business information, operational contact information, true ownership information, and debt capacity information.

Images