Global optimization method for auxiliary transportation in coal mines and computer-readable storage medium

Abstract

The present application provides a global optimization method and computer-readable storage medium for auxiliary transportation of coal mines. The method includes: obtaining approval process information and transportation requirements, the approval process information is information related to the approval order of the work order of the transportation process; according to the approval process information, optimizing the approval order of the transportation process to obtain an optimized approval order; approving the transportation requirements according to the optimized approval order to obtain an approval result, and when the approval result indicates that the approval is passed, optimizing the transportation process according to the transportation requirements to obtain an optimized transportation process. In this solution, the approval order is continuously optimized, which can simplify the approval process, and can be approved in an intelligent manner. The resource scheduling of the transportation process can also be optimized, that is, the entire process of auxiliary transportation of coal mines is globally optimized, thereby improving the efficiency of auxiliary transportation in mines.

Description

Technical Field

[0001] The present application relates to the technical field of coal mine auxiliary transportation, and in particular to a global optimization method, device, computer-readable storage medium and electronic device for coal mine auxiliary transportation. Background Art

[0002] Currently, the management models for auxiliary transportation in coal mines vary widely, both domestically and internationally. The efficiency of control processes and transportation route planning is still limited by the complexity of regulations and the management level of managers. This is compounded by the wide disparity in demand for auxiliary transportation services, complex approval processes, limited transportation resources, and the difficulty of safe operation management. This results in low efficiency in auxiliary transportation systems, further restricting the management efficiency, economic benefits, and safety production control capabilities of coal mining enterprises. Therefore, in current solutions, due to the complex approval process and manual scheduling of resources such as vehicles and personnel, the efficiency of auxiliary transportation in coal mines is low. Summary of the Invention

[0003] The main purpose of this application is to provide a global optimization method, device, computer-readable storage medium and electronic device for coal mine auxiliary transportation to solve the problem of low transportation efficiency of coal mines in the prior art.

[0004] According to one aspect of an embodiment of the present invention, a global optimization method for auxiliary transportation in coal mines is provided, comprising: obtaining approval process information and transportation requirements, wherein the approval process information is information related to the approval order of work orders in the transportation process, and the transportation requirements refer to transportation equipment requirements or transportation personnel requirements; optimizing the approval order of the transportation process according to the approval process information to obtain an optimized approval order; approving the transportation requirements according to the optimized approval order to obtain an approval result, and when the approval result indicates that the approval is passed, optimizing the transportation process according to the transportation requirements to obtain an optimized transportation process.

[0005] Optionally, there are multiple approval sequences, and the approval sequence of the transportation process is optimized according to the approval process information to obtain an optimized approval sequence, including: initializing the approval sequence, classifying the approval sequence according to the approval process information to obtain a classified approval sequence; selecting multiple approval sequences corresponding to the target category according to the transportation demand; selecting one approval sequence from the multiple approval sequences as the target approval sequence, and optimizing the approval nodes of the target approval sequence to obtain an optimized target approval sequence, wherein the approval sequence includes at least one approval node.

[0006] Optionally, optimizing the approval nodes of the target approval sequence to obtain an optimized target approval sequence includes: obtaining first approval information corresponding to the approval nodes, the first approval information at least including historical approval records; and deleting some approval nodes in the target approval sequence based on the first approval information to obtain the optimized target approval sequence.

[0007] Optionally, optimizing the approval nodes of the target approval sequence to obtain an optimized target approval sequence includes: obtaining second approval information corresponding to the approval nodes, the second approval information including at least user feedback data, historical approval records, and approval time; and replacing some approval nodes in the target approval sequence according to the second approval information to obtain the optimized target approval sequence.

[0008] Optionally, the transportation process is optimized according to the transportation demand to obtain an optimized transportation process, including: optimizing the site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency tasks of the transportation process based on the transportation demand.

[0009] Optionally, the site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency tasks of the transportation process are optimized, including: obtaining site-related information, the site-related information including at least one of the following: coal mine tunnel topology, miner personnel information, and personnel operation distribution in coal mine tunnels; optimizing the departure time of the site and the vehicle based on the site-related information; obtaining vehicle order-related information, the vehicle order-related information including at least one of the following: vehicle waiting time, vehicle cost, and vehicle allocation order; obtaining vehicle-related information, the vehicle-related information including at least one of the following: vehicle type and load capacity; optimizing the vehicle scheduling process based on the vehicle order-related information and the vehicle-related information; obtaining driver-related information, the driver-related information including at least one of the following: accumulated mileage, accumulated working hours, driving qualifications, and number of dispatches; optimizing the driver scheduling process based on the driver-related information.

[0010] Optionally, the site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency tasks of the transportation process are optimized, and also includes: obtaining route-related information, the route-related information including at least one of the following: road conditions of underground coal mine tunnels, tunnel geometric representation, map information of underground coal mine tunnels, environmental parameter change information, starting point and ending point of the vehicle; optimizing the route planning process based on the route-related information; obtaining traffic signal information-related information, the traffic signal information-related information including at least one of the following: vehicle distribution, vehicle speed, vehicle type, position sensing information corresponding to the vehicle, road conditions of coal mine tunnels; optimizing the traffic signal information configuration process based on the traffic signal information-related information; obtaining the priority order of multiple dispatched tasks, and when an emergency task is received, setting the priority of the emergency task to the highest priority.

[0011] According to another aspect of an embodiment of the present invention, a global optimization device for auxiliary transportation of coal mines is also provided, including: an acquisition unit, used to obtain approval process information and transportation requirements, the approval process information is information related to the approval order of work orders in the transportation process, and the transportation requirements refer to transportation equipment requirements or transportation personnel requirements; a first optimization unit, used to optimize the approval order of the transportation process according to the approval process information to obtain an optimized approval order; a second optimization unit, used to approve the transportation requirements according to the optimized approval order to obtain an approval result, and when the approval result indicates that the approval is passed, optimize the transportation process according to the transportation requirements to obtain an optimized transportation process.

[0012] According to yet another aspect of the embodiments of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein the program executes any one of the methods described above.

[0013] According to another aspect of an embodiment of the present invention, an electronic device is also provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of the methods described.

[0014] In an embodiment of the present invention, approval process information and transportation requirements are first obtained. Then, based on the approval process information, the approval order of the transportation process is optimized to obtain an optimized approval order. Finally, the transportation requirements are approved according to the optimized approval order to obtain an approval result. If the approval result indicates approval, the transportation process is optimized based on the transportation requirements to obtain an optimized transportation process. In this solution, continuous optimization of the approval order can simplify the approval process, enable intelligent approval, and optimize resource scheduling during the transportation process. In other words, the entire process of coal mine auxiliary transportation is globally optimized, thereby improving the efficiency of mine auxiliary transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings that constitute part of this application are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation on this application. In the drawings:

[0016] Figure 1 A schematic flow chart of a global optimization method for auxiliary transportation of coal mines according to an embodiment of the present application is shown;

[0017] Figure 2 Shows a schematic diagram of the architecture of the machine learning algorithm system;

[0018] Figure 3 A schematic diagram of the process of optimizing the approval sequence is shown;

[0019] Figure 4 A flow chart of a single-node approval opinion recommendation algorithm is shown;

[0020] Figure 5 A schematic diagram of a process for optimizing the auxiliary transportation process is shown;

[0021] Figure 6 A structural schematic diagram of a global optimization device for auxiliary transportation of coal mines according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0022] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0023] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] It should be understood that when an element (such as a layer, film, region, or substrate) is described as being "on" another element, the element may be directly on the other element or intervening elements may be present. Moreover, in the specification and claims, when it is described that an element is "connected to" another element, the element may be "directly connected to" the other element or "connected to" the other element through a third element.

[0026] For ease of description, some nouns or terms involved in the embodiments of the present application are explained below:

[0027] Optical Character Recognition: refers to the automatic recognition of text content in an image.

[0028] K-Nearest Neighbor (KNN): A machine learning classification algorithm. The idea of this method is: in the feature space, if most of the k nearest samples near a sample belong to a certain category, then the sample also belongs to this category.

[0029] Natural Language Processing (NLP) is a discipline and technology that uses computer technology to analyze, understand and process natural language.

[0030] Genetic Algorithm: Genetic algorithm is a search algorithm used to solve optimization problems in computational mathematics and is a type of evolutionary algorithm.

[0031] K-means clustering algorithm: It is an iterative clustering analysis algorithm.

[0032] As traditional information technology software, the coal mine auxiliary transportation management system encounters mismatches between its user processes and business processes during its rollout and operation at different coal mines. This often requires extensive business research, consulting, and custom development by a team of software experts. In this business sector, no research has yet explored using big data and artificial intelligence to address these process optimization issues. Related literature primarily focuses on finance, procurement, and administrative approval. Wang Jianjun et al. used the AHP-BP neural network method to construct a comprehensive scientific evaluation system for measuring internal control effectiveness, generating corresponding evaluation decision rules. They proposed a technical approach using RPA technology to automate procurement contract reading, invoice image recognition, and bill matching. Researcher Hu Sainan employed the K-means algorithm for hierarchical classification in the procurement approval process, reducing the ambiguity of human experience and the subjectivity of audit evaluations. Li Shuangqing proposed a strategy for building an intelligent real estate registration approval system, implementing internet information services and intranet approval functions, effectively improving the efficiency of real estate registration. Sun Xiaodong developed an intelligent approval system for financial reimbursement documents, providing decision support for approvers and improving approval efficiency and quality.

[0033] As for the resource scheduling link in the auxiliary transportation process, the actual business still mainly adopts the model of monitoring system assisted manual scheduling. Currently, among the patents issued for the resource and operation management of coal mine auxiliary transportation systems: Du Lei has developed a mining vehicle transportation monitoring and scheduling system that includes an on-board unit, a tunnel unit, and a scheduling unit to solve the problem of poor scheduling timeliness due to the inability of dispatchers to timely grasp the status of the system vehicles. Yang Peng has developed a technical solution based on vehicle operation dynamics, parameter receivers, and signal indicators to achieve local coexistence and avoidance of mine vehicles; in the academic field of vehicle resource scheduling research, Li Mengyu and others used CPNTools to establish a hierarchical color Petri net model of the coal mine auxiliary transportation vehicle scheduling system, and analyzed the impact of different scheduling strategies on auxiliary transportation efficiency through simulation. Zhang Lixiang designed a coal mine auxiliary transportation intelligent scheduling platform based on UWB positioning technology, which includes real-time monitoring, statistical analysis, alarm, operation scheduling and other functions to digitally manage the auxiliary transportation equipment and materials of the entire mine.

[0034] Some researchers have explored the application of AI in transactional scenarios such as finance, procurement, and administration, but these efforts are often limited to single-point intelligent applications such as text recognition, supplier evaluation, and process automation. Based on current literature research, no comprehensive technical content has been found that encompasses the entire lifecycle of the system approval process, from initialization and establishment to evaluation system construction and iterative optimization. Furthermore, the industry-specific nature of coal mine auxiliary transportation systems, with their numerous components and complex business processes, necessitates adaptive improvement and adjustment of the approval process to optimize back-end resource scheduling. However, relatively few relevant research papers and patents exist.

[0035] In the context of optimizing resource scheduling processes, current auxiliary transportation management systems still primarily focus on basic functions such as personnel positioning, vehicle monitoring, business management, and statistical analysis, lacking intelligent resource scheduling engines driven by next-generation information technologies such as big data and artificial intelligence. Academic research primarily relies on discrete events, using Petri networks and commercial software to establish simplified simulation environments for data analysis and vehicle scheduling optimization. However, in real-world production scenarios, coal mine auxiliary transportation systems encompass different types of vehicle needs, including planned, temporary, and emergency needs. These systems cater to numerous demand units, including mining, excavation, machinery, transportation, and access, and involve multiple elements, including routes, personnel, vehicles, traffic signal information, environmental safety, and stations. These elements are highly coupled, and current research lacks comprehensive, real-time resource integration, safety assessment, and scheduling capabilities.

[0036] As mentioned in the background technology, the transportation efficiency of coal mines in the existing technology is low. In order to solve the above problem, in a typical embodiment of the present application, a global optimization method, device, computer-readable storage medium and electronic device for coal mine auxiliary transportation are provided.

[0037] According to an embodiment of the present application, a global optimization method for auxiliary transportation in coal mines is provided.

[0038] Figure 1 Flowchart of the global optimization method of coal mine auxiliary transportation according to the embodiment of the present application. Figure 1 As shown, the method includes the following steps:

[0039] Step S101: Acquire approval process information and transportation requirements. The approval process information is information related to the approval order of work orders in the transportation process. The transportation requirements refer to transportation equipment requirements or transportation personnel requirements.

[0040] In the above step S101, approval process information and transportation requirements can be obtained, and then subsequent optimization processes can be carried out based on the approval process information and transportation requirements.

[0041] Step S102: Optimize the approval sequence of the transportation process based on the above approval process information to obtain an optimized approval sequence;

[0042] In the above step S102, the approval sequence of the transportation process can be optimized through the approval process information, and the approval sequence of mine auxiliary transportation can be continuously simplified and optimized, avoiding a relatively complicated and unscientific approval process.

[0043] Step S103: Approve the transportation demand in accordance with the optimized approval order to obtain an approval result. If the approval result indicates approval, optimize the transportation process according to the transportation demand to obtain an optimized transportation process.

[0044] In the above step S103, the entire transportation process can be globally optimized, which can alleviate the problems of low efficiency of manual scheduling resource utilization, unreasonable decision-making, and difficulty in ensuring safety mechanisms, thereby further effectively improving the auxiliary transportation efficiency and safety management level of the mine.

[0045] In the above method, approval process information and transportation requirements are first obtained. Then, based on the approval process information, the approval order of the transportation process is optimized to obtain an optimized approval order. Finally, the transportation requirements are approved according to the optimized approval order to obtain an approval result. If the approval result indicates that the approval is passed, the transportation process is optimized based on the transportation requirements to obtain an optimized transportation process. In this solution, continuous optimization of the approval order can simplify the approval process, enable intelligent approval, and optimize the resource scheduling of the transportation process. In other words, the entire process of coal mine auxiliary transportation is globally optimized, thereby improving the efficiency of mine auxiliary transportation.

[0046] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0047] This solution addresses the issues described in the aforementioned background technology by researching and designing a machine learning algorithm system for the approval process and operational scheduling of coal mine auxiliary transportation systems. This technical solution is based on core AI technologies such as big data storage and computing, OCR text recognition, NLP natural language processing, machine learning, operational optimization, and reinforcement learning. It builds an AI algorithm system comprised of two intelligent engines: AI approval process intelligent optimization and AI resource scheduling optimization, and over ten algorithm models in total. This system aims to continuously improve the management and operational efficiency of coal mine auxiliary transportation systems, helping coal mining enterprises achieve their goals of safe, green, intelligent, and efficient enterprise development.

[0048] This solution aims to optimize the transportation approval process and transportation process by developing an artificial intelligence model system. Based on the current management process, it can automatically generate and continuously optimize scientific and reasonable management and transportation processes through big data analysis. It mainly solves the problems of management failure and timeliness caused by the unreasonable approval process of auxiliary transportation in mines; and the problems of tight and wasted transportation equipment resources caused by low efficiency in the transportation of materials and personnel. Moreover, this model system is based on the coal industry and has wide applicability and economic value for auxiliary transportation control processes of various management models. This solution is a necessary part of the intelligent construction of the coal industry and is of great significance to the optimization and reconstruction of enterprise management processes and the reduction of costs and increase of efficiency.

[0049] Specifically, the machine learning algorithm process of the machine learning algorithm system is as follows: Figure 2 As shown, driven by dual intelligent engines—an AI approval process optimization engine and an AI resource scheduling optimization engine—the system has built an intelligent model system covering more than 10 scenarios, including approval process initialization and optimization, vehicle scheduling, route planning, and driver scheduling, targeting multiple user roles, including senior coal mine managers, vehicle demanders, vehicle approvers (for example, production departments, safety management departments, electromechanical departments, and other business departments), transportation equipment dispatchers, and task executors (for example, drivers and warehouse managers). This system utilizes AI technology to achieve better management and operational results throughout the life cycle of the auxiliary transportation system. The approval process includes the approval sequence. Figure 2 The functions of each engine and each user role in are as follows:

[0050] Senior management of coal mines: Approve the initialized approval process, select the appropriate approval process from multiple approval processes, and after the approval process is selected, the specific vehicle approval party will approve it.

[0051] Vehicle demanders: input vehicle demand (transportation demand) and submit an order (order for vehicle demand).

[0052] Vehicle approval party: There are multiple approval processes, including approval process 1, approval process 2...approval process N, and the approval process is approved. If the approval result indicates that the approval is passed, the AI resource scheduling optimization engine will optimize the transportation process. If the approval result indicates that it is not passed, the vehicle demander will be notified.

[0053] AI approval process optimization engine: obtains approval process information, which may include information on paper documents and approval rules and regulations, and initializes the approval process by the process initialization model; the single-point intelligent approval model based on case reasoning makes intelligent recommendations on the approval opinions of the vehicle approver, and optimizes the approval process of the vehicle approver based on stored approval records, user feedback records and approval process optimization model.

[0054] AI resource scheduling optimization engine: Obtains data on maps, lane geometry, vehicles, personnel and materials, and the environment, and optimizes the transportation process based on site configuration models, vehicle scheduling models, driver scheduling models, path optimization models, traffic signal information setting models, and emergency task models.

[0055] Transport equipment dispatcher: After optimizing the approval process and the transport process, there is no need for the transport equipment dispatcher to manually dispatch resources. Resources can be dispatched automatically, and the transport equipment dispatcher can monitor the resource scheduling.

[0056] Task executor: Execute the task according to the approval process, transportation process and order requirements. After the task is completed, notify the vehicle demander that the vehicle demand has been met.

[0057] In one embodiment of the present application, there are multiple approval sequences. The approval sequence of the transportation process is optimized based on the approval process information to obtain the optimized approval sequence, which specifically includes the following steps:

[0058] Step S201, initializing the approval order, classifying the approval order according to the approval process information, and obtaining a classified approval order;

[0059] Specifically, a process initialization model can be used to initialize the approval sequence. If there is no digital approval process data, this solution uses OCR technology to perform text detection and automatic text recognition on paper documents; after the recognized text data is pre-processed through numerical vector representation, word segmentation, normalization and missing filling using NLP technology, historical approval data that meets the standard data structure is obtained; considering the data distribution, this solution can use the kernel K-means unsupervised algorithm to perform cluster analysis on the approval process data (which may include advance vehicle needs, temporary vehicle needs, and fixed vehicle needs), and obtain more accurate clustering results by introducing kernel functions, thereby dividing the approval process into approval processes for different vehicle-using departments and different vehicle needs. The generated initialized approval process will take effect after review and approval by senior management of the coal mine.

[0060] Step S202: selecting multiple approval orders corresponding to the target category according to the above transportation requirements;

[0061] Step S203: Select one of the approval sequences as the target approval sequence from the multiple approval sequences, and optimize the approval nodes of the target approval sequence to obtain an optimized target approval sequence, wherein the approval sequence includes at least one of the approval nodes.

[0062] Specifically, based on user feedback records, historical approval records, approval time and other data, we can use heuristic algorithms such as genetic algorithms to optimize the target approval order. We can optimize the approval departments, approval nodes and approvers, and finally obtain the optimized standard target approval order. The process of this algorithm is as follows Figure 3 As shown:

[0063] 1) Population initialization and agent encoding: In the coal mine auxiliary transportation system approval process scenario, the encoding represents the department and specific approver in the approval process. This solution uses floating-point encoding to ensure that the gene value is within a given range. The gene value range is set to 0 to 1, where 0 indicates that the approval layer is not required, and 1 indicates that all approvers in the approval layer are included in the approval process.

[0064] 2) Establish a fitness function: The fitness function of this solution needs to ensure that the approval process goes through as few approval departments, approval nodes, and approvers as possible, while at the same time ensuring that the approval process has the greatest possible security value (the credit evaluation value of each vehicle demander is set by the key person in charge of the coal mine).

[0065] 3) Selection: This determines how individuals are selected from the parent population to be passed down to the next generation. The selection process determines which individuals will undergo recombination or crossover, as well as the number of offspring individuals produced by the selected individuals. This solution preserves efficient approval processes and maximizes the number of offspring with higher processing efficiency and adaptability.

[0066] 4) Crossover and Mutation Operations: In the coal mine assisted transportation scenario, a crossover operation is defined as the gene exchange between different parent agents, resulting in a new child agent by exchanging the approval nodes in the gene sequence. Mutation operations in the coal mine assisted transportation scenario can be divided into two aspects: first, replacing the approval node, i.e., switching the current approval department to another; second, replacing the current approver with another approver.

[0067] 5) Determine whether the termination condition is met. If so, end the optimization process of the approval order. If not, re-establish the fitness function for calculation.

[0068] In order to further optimize the approval nodes of the target approval sequence and shorten the target approval sequence, in a specific embodiment of the present application, the approval nodes of the target approval sequence are optimized to obtain the optimized target approval sequence, which specifically includes the following steps:

[0069] Step S2031: Acquire first approval information corresponding to the approval node, where the first approval information at least includes historical approval records;

[0070] Step S2032: based on the first approval information, some approval nodes in the target approval sequence are deleted to obtain an optimized target approval sequence.

[0071] Specifically, the original approval order is approval node A, approval node B, approval node C, approval node D, and approval node E. Since approval node C does not participate in the approval every time, approval node C can be deleted from the approval order. The optimized approval order is approval node A, approval node B, approval node D, and approval node E.

[0072] In the above steps S2031 to S2032, by deleting some approval nodes in the target approval sequence, the target approval sequence can be shortened, so that the approval time can be shortened by performing approval according to the optimized target approval sequence.

[0073] To further optimize the approval nodes of the target approval sequence, some approval nodes in the target approval sequence may be replaced. In another specific embodiment of the present application, the approval nodes of the target approval sequence are optimized to obtain an optimized target approval sequence, specifically comprising the following steps:

[0074] Step S2033: Acquire the second approval information corresponding to the approval node. The second approval information at least includes user feedback data, historical approval records, and approval time.

[0075] Step S2034: According to the second approval information, some approval nodes in the target approval sequence are replaced to obtain an optimized target approval sequence.

[0076] The above steps are not limited to deleting and replacing approval nodes, but also can add necessary safety approval nodes according to the actual operating environment and tasks of the auxiliary transportation equipment.

[0077] In the above steps S2033 to S2034, by replacing some approval nodes in the target approval sequence, approval nodes with higher approval efficiency can be used for approval, so that approval according to the optimized target approval sequence can shorten the approval time.

[0078] In practical applications, in addition to the aforementioned optimization methods of deleting and replacing approval nodes, a single-point intelligent approval model based on case-based reasoning can also be used for optimization. Specifically, this single-point intelligent approval process model, leveraging big data and artificial intelligence technologies, addresses the issue of unnecessary, burdensome, and inefficient manual approval for similar and reasonable vehicle requests, thereby reducing approval process time. The core concept of this solution is to continuously accumulate historical information on the auxiliary transportation system's approval process and store it in a case library in the format of <case description, case approval description>. This case library utilizes Hadoop as its underlying development platform, employing technologies such as the HBase distributed database, the MapReduce distributed computing framework, and Solr distributed full-text search to store, compute, and search approval cases.

[0079] When a new approval requirement arises, the case database can be used to find the case approval result that is most similar to the current approval case according to the set case similarity formula, and the system can be used to recommend approval opinions to reduce the number of manual approval nodes. This technical solution mainly uses four key processes to solve the problem: case representation, retrieval, reuse and correction, and evaluation and learning. The brief workflow is as follows: Figure 4 As shown:

[0080] 1) Approval case representation: When obtaining approval requirements, natural language processing and other technologies are used to extract keywords and features from the newly generated vehicle use requirements, and then represent them as a formatted approval case (i.e., a new case) containing detailed information such as the applicant and vehicle use application.

[0081] Case representation uses a certain data structure to describe the characteristics of the case: the characteristics of the approval case in this solution include three categories: demand source characteristics, order characteristics, and approval characteristics, which can be described as: Case = {[personnel using the vehicle, department using the vehicle], [time of vehicle use, purpose of vehicle use, [material requirement 1, material requirement 2]], personnel requirements], [approval node 1, approval opinion 1, approval time 1, approval notes 1], [approval node 2, approval opinion 2, approval time 2, approval notes 2], [approval node n, approval opinion n, approval time n, approval notes n]]}.

[0082] 2) Approval Case Retrieval: Approval case retrieval calculates the similarity between the new approval requirement and completed approval cases in the case library to obtain cases with high similarity. Similarity calculation methods can be used for different data types, such as continuous numerical attributes, ordered attributes, symbolic categorical attributes, and character attributes. A hybrid calculation method using cosine similarity and Jaccard similarity coefficient can be used. Based on the similarity calculation results, algorithms such as nearest neighbor and K-nearest neighbor can be used to obtain retrieval results.

[0083] The above similarity calculation process is:

[0084] As shown in the above data structure, there are different methods for ordering, mainly including text description, logical classification and numerical representation. Among them:

[0085] The text description similarity d1 can be calculated using vocabulary similarity algorithms such as TF-IDF, Doc2Vec, and Word2Vec;

[0086] The logical classification similarity d2 can be calculated using the following algorithm formula:

[0087]

[0088] Where xi represents the current case, yi represents the historical case, and n represents the number of relevant information related to the case. The relevant information includes at least one of the following: station-related information, car order-related information, vehicle-related information, driver-related information, route-related information, and traffic signal information-related information. The similarity between the current case and the historical case can be calculated through the logical classification similarity d2;

[0089] The numerical representation of similarity d3 can be calculated using the following algorithm formula:

[0090]

[0091] ωi represents a predefined weight value, which is the weight value of the difference between the current case and the historical case, αi=xi max -xi min ,xi max represents the maximum value of relevant information related to the case, xi min Represents the minimum value of relevant information related to the case, and this calculation process is used to achieve parameter normalization.

[0092] In this scheme, the similarity of approval cases refers to the similarity calculated for all attributes of the entire case. Therefore, the above algorithm can be used to calculate the similarity of each type separately and then perform weighted summation (each weight value can be assigned manually or automatically adjusted through manual evaluation). The specific formula is: dtotal = ω1×d1+ω2×d2+ω3×d3, dtotal represents the final calculated similarity, ω1 represents the weight of the calculation result obtained by text description similarity, ω2 represents the weight of the calculation result obtained by logical classification similarity, and ω3 represents the weight of the calculation result obtained by numerical representation similarity.

[0093] 3) Reuse and manual correction of approval results: Based on the retrieval results of similar cases based on the set model, if the new approval requirements are successfully matched, the case recommendation result with the highest similarity is output and recommended for automatic execution; otherwise, manual approval and correction steps are involved, and the corrected results are verified. The corrected results are also returned to the case reasoning system as high-quality samples for iterative learning by the intelligent model.

[0094] 4) Case Evaluation and Learning: With the continuous accumulation of post-approval cases and user feedback on approval results (i.e., approval case evaluation), this technical solution adopts the AutoML method to regularly perform self-adaptation and self-learning to maintain the approval sequence, so that the process approval reasoning results can be closer to business needs and logic, greatly improving the system's reasoning capabilities; approval cases can also be stored in the approval case library.

[0095] In the above steps S201 to S203, the approval sequence is first initialized, and then the approval sequence of different categories can be optimized. This can continuously simplify and optimize the approval sequence of coal mines and avoid the approval sequence being more complicated.

[0096] For the coal mine transportation process, it includes the links of site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency task allocation. Therefore, the above links can be optimized. In another embodiment of the present application, the transportation process is optimized according to the above transportation needs to obtain the optimized transportation process, which specifically includes the following steps: based on the above transportation needs, the site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency tasks of the transportation process are optimized. In this embodiment, by optimizing the site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency tasks of the transportation process, efficient coordination and allocation of various resources in the auxiliary transportation process can be achieved.

[0097] Specifically, an AI resource scheduling optimization engine can be used to optimize the transportation process. The AI resource scheduling optimization engine is oriented towards the previously approved vehicle order requirements. Based on the collected historical / real-time data, it uses big data, artificial intelligence, operations optimization and other technologies to achieve intelligent perception and real-time global optimization of the core elements of the auxiliary transportation system, such as personnel, vehicles, materials, traffic lights, and driving routes.

[0098] In order to further optimize the transportation process more efficiently, optimization can be performed based on various relevant information of the transportation process. By obtaining multiple relevant information, the links of the transportation process can be optimized. In a specific embodiment of the present application, the site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency tasks of the transportation process are optimized, which specifically includes the following steps:

[0099] Step S301, obtaining site-related information, the site-related information including at least one of the following: coal mine roadway topology, miner information, and personnel operation distribution in the coal mine roadway;

[0100] Step S302: Optimize the station and vehicle departure time based on the station-related information;

[0101] Specifically, the station configuration model can be used to optimize the station and vehicle departure time, such as Figure 5 As shown, the station configuration model is primarily designed for personnel transport vehicles with relatively fixed schedules, routes, and stations. It aims to reduce waiting times and walking distances during peak usage periods. The model uses a mine tunnel topology, miner information, feasible pre-set stations, and personnel work distribution as its primary inputs. Using a genetic algorithm, it calculates the locations of each station over a period of time, combines these stations into a route, and provides the starting departure time for the shuttle bus route.

[0102] Step S303: obtaining information related to the vehicle order, wherein the information related to the vehicle order includes at least one of the following: vehicle waiting time, vehicle cost, and vehicle allocation order;

[0103] Step S304: obtaining vehicle-related information, wherein the vehicle-related information includes at least one of the following: vehicle type and load capacity;

[0104] Step S305: Optimizing the vehicle dispatch process based on the vehicle order information and the vehicle information;

[0105] Specifically, the vehicle scheduling optimization model can be used to optimize the vehicle scheduling process, such as Figure 5As shown in the figure, the vehicle scheduling optimization model, combined with vehicle demand priorities, uses vehicle demand (including planned and temporary use) and available vehicle information as primary inputs. It selects the optimal vehicle to perform specific transport tasks, targeting minimum vehicle empty mileage, vehicle waiting time, and vehicle cost. By splitting large demand orders into small transport work orders and then combining small demand orders into single orders, the model implements a logistics warehouse management model for vehicle scheduling that responds to multiple vehicles per order, or multiple orders per vehicle, thereby improving vehicle transport efficiency. The model's key constraints include accepting all high-priority orders, matching vehicle transport capacity with work order requirements, ensuring that different work order schedules do not conflict, and ensuring that the vehicle's mileage is no less than the transport distance.

[0106] Step S306: Obtain driver-related information, which includes at least one of the following: accumulated mileage, accumulated working hours, driving qualifications, and number of dispatches;

[0107] Step S307: Optimize the driver's dispatch process based on the above driver-related information.

[0108] Specifically, the driver scheduling model can be used to optimize the driver scheduling process, such as Figure 5 As shown, the driver scheduling model assigns drivers to each vehicle assigned to a work order in the vehicle scheduling optimization model based on demand order information and driver information (including accumulated mileage, accumulated working hours, driving qualifications, and number of dispatches). This model aims to balance each driver's recent accumulated mileage, time, and wages. It can be solved using operations optimization or heuristic algorithms, and primarily includes constraints such as matching driver and vehicle qualifications, requiring each vehicle to be driven by a human, and requiring each driver to drive a maximum of one vehicle.

[0109] In the above steps S301 to S307, the departure time of the station and the vehicle can be optimized through the station-related information, the vehicle scheduling process can be optimized through the vehicle-related information, and the driver scheduling process can be optimized through the driver-related information, thereby further realizing the efficient coordination and allocation of various resources in the auxiliary transportation process.

[0110] To further optimize the transportation process more efficiently, optimization can be performed based on various relevant information of the transportation process. By obtaining multiple relevant information, the links of the transportation process can be optimized. In another specific embodiment of the present application, the site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration, and emergency tasks of the transportation process are optimized, and the following steps are specifically included:

[0111] Step S308: Acquire path-related information, where the path-related information includes at least one of the following: road conditions of the underground coal mine tunnel, tunnel geometry, map information of the underground coal mine tunnel, environmental parameter change information, and the vehicle's starting and ending points;

[0112] Specifically, the environmental parameter change information refers to information about changes in the environment of the mine, including at least information about changes in dust, information about changes in smoke, information about whether there is water in the mine, and information about whether there is danger in the mine.

[0113] Step S309: Optimizing the path planning process based on the above path related information;

[0114] Specifically, the path optimization model can be used to optimize the route of the vehicle to perform the transportation task, such as Figure 5 As shown in the figure, assisted transport drivers often face the problem of incorrect driving routes and unclear drivable routes after changes in working faces or tunnels. The intelligent path planning model provides recommended routes for drivers based on coal mine map information, transport work order information (including pick-up and drop-off points), vehicle attribute information, and real-time status monitoring data. It can also inform drivers of subsequent driving behaviors in advance based on the vehicle route and current status. When driving on the wrong road, it can prompt and re-plan the route. Compared with road traffic planning, the path planning model has the particularity of industry scenarios and is more complex to implement. Differentiated data processing and business modeling are required for this business. The model takes the minimum driving path and driving time as its main goals, and can be solved using a variety of technologies such as the A-Star algorithm, heuristic algorithm, and reinforcement learning.

[0115] Step S310: Obtaining traffic signal information related information, wherein the traffic signal information related information includes at least one of the following: vehicle distribution, vehicle speed, vehicle type, position sensing information corresponding to the vehicle, and road conditions of coal mine tunnels;

[0116] Specifically, the position sensing information corresponding to the vehicle refers to information sensed by a coil at a position corresponding to the vehicle.

[0117] Step S311, optimizing the configuration process of traffic signal information based on the above traffic signal information related information;

[0118] Specifically, the traffic signal information setting model can be used to optimize the configuration process of traffic signal information, such as Figure 5As shown, this model achieves the following business objectives by collaboratively controlling underground traffic signal information: reducing unnecessary vehicle stops and waiting times and ensuring future driving safety; providing an alarm message and taking over traffic signal control at the intersection when the induction coil of the traffic signal information system at the intersection fails; and taking over traffic signal control at the intersection when a high-priority emergency task occurs to ensure smooth passage of emergency vehicles. This solution simulates and trains underground traffic signal scheduling using deep reinforcement learning methods. Multiple agents independently control each signal light, while introducing mutual information for collaboration, achieving real-time optimization and scheduling of underground traffic signal information settings.

[0119] Step S312: obtaining the priority order of the multiple dispatched tasks, and in the case of receiving an urgent task, setting the priority of the urgent task to the highest priority.

[0120] Specifically, if Figure 5 As shown, an emergency mission model can be used to integrate the five aforementioned models: site configuration, vehicle scheduling, driver scheduling, route planning, and traffic signal information control. For emergency missions at the security level, comprehensive coordinated control and optimization are performed to ensure the highest priority for resource allocation and the timely arrival of required vehicles and personnel. This model requires high response time, integrating historical data to predict the status of the auxiliary transportation system in future time periods. Furthermore, heuristic algorithms or reinforcement learning algorithms are used for decision optimization to accelerate the solution process and achieve rapid response.

[0121] In the above steps S306 to S312, the path planning process can be optimized through path-related information, and the traffic signal information configuration process can be optimized through traffic signal information-related information. In addition, other optimization processes can be coordinated in the event of an emergency task, thereby further realizing efficient coordination and allocation of various resources in the auxiliary transportation process.

[0122] The present application also provides a global optimization device for auxiliary coal mine transportation. It should be noted that the global optimization device for auxiliary coal mine transportation in the present application can be used to execute the global optimization method for auxiliary coal mine transportation provided in the present application. The global optimization method for auxiliary coal mine transportation provided in the present application is described below.

[0123] Figure 6 Schematic diagram of a global optimization device for auxiliary transportation of coal mines according to an embodiment of the present application. Figure 6 As shown, the device includes:

[0124] An acquisition unit 10 is configured to acquire approval process information and transportation requirements. The approval process information is information related to the approval order of work orders in the transportation process. The transportation requirements refer to transportation equipment requirements or transportation personnel requirements.

[0125] The acquisition unit mentioned above can acquire the approval process information and transportation requirements, and then perform subsequent optimization processes based on the approval process information and transportation requirements.

[0126] The first optimization unit 20 is used to optimize the approval sequence of the transportation process according to the above-mentioned approval process information to obtain an optimized approval sequence;

[0127] The above-mentioned first optimization unit can optimize the approval sequence of the transportation process through the approval process information, and then can continuously simplify and optimize the approval sequence of mine auxiliary transportation, avoiding a more complicated and unscientific approval process.

[0128] The second optimization unit 30 is used to approve the above-mentioned transportation requirements according to the optimized approval order to obtain an approval result, and when the above-mentioned approval result indicates that the approval is passed, optimize the transportation process according to the above-mentioned transportation requirements to obtain an optimized transportation process.

[0129] The above-mentioned second optimization unit can globally optimize the entire transportation process, alleviate the problems of low efficiency of manual scheduling resource utilization, unreasonable decision-making, and difficulty in ensuring safety mechanisms, and further effectively improve the auxiliary transportation efficiency and safety management level of the mine.

[0130] In the above-mentioned device, the acquisition unit obtains approval process information and transportation requirements. The first optimization unit optimizes the approval order of the transportation process based on the approval process information to obtain an optimized approval order. The second optimization unit approves the transportation requirements according to the optimized approval order to obtain an approval result. If the approval result indicates that the approval is passed, the transportation process is optimized based on the transportation requirements to obtain an optimized transportation process. In this solution, continuous optimization of the approval order can simplify the approval process, and approval can be performed in an intelligent manner. The resource scheduling of the transportation process can also be optimized, that is, the entire process of coal mine auxiliary transportation is globally optimized, thereby improving the efficiency of mine auxiliary transportation.

[0131] This solution addresses the issues described in the aforementioned background technology by researching and designing a machine learning algorithm system for the approval process and operational scheduling of coal mine auxiliary transportation systems. This technical solution is based on core AI technologies such as big data storage and computing, OCR text recognition, NLP natural language processing, machine learning, operational optimization, and reinforcement learning. It builds an AI algorithm system comprised of two intelligent engines: AI approval process intelligent optimization and AI resource scheduling optimization, and over ten algorithm models in total. This system aims to continuously improve the management and operational efficiency of coal mine auxiliary transportation systems, helping coal mining enterprises achieve their goals of safe, green, intelligent, and efficient enterprise development.

[0132] This solution aims to optimize the transportation approval process and transportation process by developing an artificial intelligence model system. Based on the current management process, it can automatically generate and continuously optimize scientific and reasonable management and transportation processes through big data analysis. It mainly solves the problems of management failure and timeliness caused by the unreasonable approval process of auxiliary transportation in mines; and the problems of tight and wasted transportation equipment resources caused by low efficiency in the transportation of materials and personnel. Moreover, this model system is based on the coal industry and has wide applicability and economic value for auxiliary transportation control processes of various management models. This solution is a necessary part of the intelligent construction of the coal industry and is of great significance to the optimization and reconstruction of enterprise management processes and the reduction of costs and increase of efficiency.

[0133] Specifically, the machine learning algorithm process of the machine learning algorithm system is as follows: Figure 2 As shown, driven by dual intelligent engines—an AI approval process optimization engine and an AI resource scheduling optimization engine—the system has built an intelligent model system covering more than 10 scenarios, including approval process initialization and optimization, vehicle scheduling, route planning, and driver scheduling, targeting multiple user roles, including senior coal mine managers, vehicle demanders, vehicle approvers (for example, production departments, safety management departments, electromechanical departments, and other business departments), transportation equipment dispatchers, and task executors (for example, drivers and warehouse managers). This system utilizes AI technology to achieve better management and operational results throughout the life cycle of the auxiliary transportation system. The approval process includes the approval sequence. Figure 2 The functions of each engine and each user role in are as follows:

[0134] Senior management of coal mines: Approve the initialized approval process, select the appropriate approval process from multiple approval processes, and after the approval process is selected, the specific vehicle approval party will approve it.

[0135] Vehicle demanders: input vehicle demand (transportation demand) and submit an order (order for vehicle demand).

[0136] Vehicle approval party: There are multiple approval processes, including approval process 1, approval process 2...approval process N, and the approval process is approved. If the approval result indicates that the approval is passed, the AI resource scheduling optimization engine will optimize the transportation process. If the approval result indicates that it is not passed, the vehicle demander will be notified.

[0137] AI approval process optimization engine: obtains approval process information, which may include information on paper documents and approval rules and regulations, and initializes the approval process by the process initialization model; the single-point intelligent approval model based on case reasoning makes intelligent recommendations on the approval opinions of the vehicle approver, and optimizes the approval process of the vehicle approver based on stored approval records, user feedback records and approval process optimization model.

[0138] AI resource scheduling optimization engine: Obtains data on maps, lane geometry, vehicles, personnel and materials, and the environment, and optimizes the transportation process based on site configuration models, vehicle scheduling models, driver scheduling models, path optimization models, traffic signal information setting models, and emergency task models.

[0139] Transport equipment dispatcher: After optimizing the approval process and the transport process, there is no need for the transport equipment dispatcher to manually dispatch resources. Resources can be dispatched automatically, and the transport equipment dispatcher can monitor the resource scheduling.

[0140] Task executor: Execute the task according to the approval process, transportation process and order requirements. After the task is completed, notify the vehicle demander that the vehicle demand has been met.

[0141] In one embodiment of the present application, there are multiple approval orders. The first optimization unit includes a processing module, a selection module, and a first optimization module. The functions of each module are as follows:

[0142] A processing module is used to initialize the approval sequence, classify the approval sequence according to the approval process information, and obtain the classified approval sequence;

[0143] Specifically, a process initialization model can be used to initialize the approval sequence. If there is no digital approval process data, this solution uses OCR technology to perform text detection and automatic text recognition on paper documents; after the recognized text data is pre-processed through numerical vector representation, word segmentation, normalization and missing filling using NLP technology, historical approval data that meets the standard data structure is obtained; considering the data distribution, this solution can use the kernel K-means unsupervised algorithm to perform cluster analysis on the approval process data (which may include advance vehicle needs, temporary vehicle needs, and fixed vehicle needs), and obtain more accurate clustering results by introducing kernel functions, thereby dividing the approval process into approval processes for different vehicle-using departments and different vehicle needs. The generated initialized approval process will take effect after being reviewed and approved by senior coal mine managers.

[0144] A selection module is used to select multiple approval sequences corresponding to the target category according to the above transportation requirements;

[0145] The first optimization module is used to select one of the above approval sequences from the multiple approval sequences as a target approval sequence, and optimize the approval nodes of the above target approval sequence to obtain an optimized target approval sequence, wherein the above approval sequence includes at least one of the above approval nodes.

[0146] Specifically, based on user feedback records, historical approval records, approval time and other data, we can use heuristic algorithms such as genetic algorithms to optimize the target approval order. We can optimize the approval departments, approval nodes and approvers, and finally obtain the optimized standard target approval order. The process of this algorithm is as follows Figure 3 As shown:

[0147] 1) Population initialization and agent encoding: In the coal mine auxiliary transportation system approval process scenario, the encoding represents the department and specific approver in the approval process. This solution uses floating-point encoding to ensure that the gene value is within a given range. The gene value range is set to 0 to 1, where 0 indicates that the approval layer is not required, and 1 indicates that all approvers in the approval layer are included in the approval process.

[0148] 2) Establish a fitness function: The fitness function of this solution needs to ensure that the approval process goes through as few approval departments, approval nodes, and approvers as possible, while at the same time ensuring that the approval process has the greatest possible security value (the credit evaluation value of each vehicle demander is set by the key person in charge of the coal mine).

[0149] 3) Selection: This determines how individuals are selected from the parent population to be passed down to the next generation. The selection process determines which individuals will undergo recombination or crossover, as well as the number of offspring individuals produced by the selected individuals. This solution preserves efficient approval processes and maximizes the number of offspring with higher processing efficiency and adaptability.

[0150] 4) Crossover and Mutation Operations: In the coal mine assisted transportation scenario, a crossover operation is defined as the gene exchange between different parent agents, resulting in a new child agent by exchanging the approval nodes in the gene sequence. Mutation operations in the coal mine assisted transportation scenario can be divided into two aspects: first, replacing the approval node, i.e., switching the current approval department to another; second, replacing the current approver with another approver.

[0151] 5) Determine whether the termination condition is met. If so, end the optimization process of the approval order. If not, re-establish the fitness function for calculation.

[0152] To further optimize the approval nodes of the target approval sequence and shorten the target approval sequence, in a specific embodiment of the present application, the first optimization module includes a first acquisition submodule and a deletion submodule, and the functions of each submodule are as follows:

[0153] A first acquisition submodule is configured to acquire first approval information corresponding to the approval node, wherein the first approval information at least includes historical approval records;

[0154] The deletion submodule is used to delete some approval nodes in the target approval sequence according to the first approval information to obtain an optimized target approval sequence.

[0155] Specifically, the original approval order is approval node A, approval node B, approval node C, approval node D, and approval node E. Since approval node C does not participate in the approval every time, approval node C can be deleted from the approval order. The optimized approval order is approval node A, approval node B, approval node D, and approval node E.

[0156] The first acquisition submodule and the deletion submodule can shorten the target approval sequence by deleting some approval nodes in the target approval sequence. In this way, the approval time can be shortened by performing approval according to the optimized target approval sequence.

[0157] To further optimize the approval nodes of the target approval sequence, some approval nodes in the target approval sequence may be replaced. In another specific embodiment of the present application, the first optimization module includes a second acquisition submodule and a replacement submodule, and the functions of each submodule are as follows:

[0158] A second acquisition submodule is configured to acquire second approval information corresponding to the approval node, wherein the second approval information includes at least user feedback data, historical approval records, and approval time;

[0159] The replacement submodule is used to replace some approval nodes in the target approval sequence according to the second approval information to obtain an optimized target approval sequence.

[0160] The above steps are not limited to deleting and replacing approval nodes, but also can add necessary safety approval nodes according to the actual operating environment and tasks of the auxiliary transportation equipment.

[0161] The second acquisition submodule and the replacement submodule mentioned above can use approval nodes with higher approval efficiency for approval by replacing some approval nodes in the target approval sequence. In this way, approval according to the optimized target approval sequence can shorten the approval time.

[0162] In practical applications, in addition to the aforementioned optimization methods of deleting and replacing approval nodes, a single-point intelligent approval model based on case-based reasoning can also be used for optimization. Specifically, this single-point intelligent approval process model, leveraging big data and artificial intelligence technologies, addresses the issue of unnecessary, burdensome, and inefficient manual approval for similar and reasonable vehicle requests, thereby reducing approval process time. The core concept of this solution is to continuously accumulate historical information on the auxiliary transportation system's approval process and store it in a case library in the format of <case description, case approval description>. This case library utilizes Hadoop as its underlying development platform, employing technologies such as the HBase distributed database, the MapReduce distributed computing framework, and Solr distributed full-text search to store, compute, and search approval cases.

[0163] When a new approval requirement arises, the case database can be used to find the case approval result that is most similar to the current approval case according to the set case similarity formula, and the system can be used to recommend approval opinions to reduce the number of manual approval nodes. This technical solution mainly uses four key processes to solve the problem: case representation, retrieval, reuse and correction, and evaluation and learning. The brief workflow is as follows: Figure 4 As shown: 1) Approval case representation: When the approval requirements are obtained, natural language processing and other technologies are used to extract keywords and features from the newly generated car use requirements, and represent them as a formatted approval case (i.e., a new case), which includes detailed information such as the applicant and the car use application.

[0164] Case representation uses a certain data structure to describe the characteristics of the case: the characteristics of the approval case in this solution include three categories: demand source characteristics, order characteristics, and approval characteristics, which can be described as: Case = {[personnel using the vehicle, department using the vehicle], [time of vehicle use, purpose of vehicle use, [material requirement 1, material requirement 2]], personnel requirements], [approval node 1, approval opinion 1, approval time 1, approval notes 1], [approval node 2, approval opinion 2, approval time 2, approval notes 2], [approval node n, approval opinion n, approval time n, approval notes n]]}.

[0165] 2) Approval Case Retrieval: Approval case retrieval calculates the similarity between the new approval requirement and completed approval cases in the case library to obtain cases with high similarity. Similarity calculation methods can be used for different data types, such as continuous numerical attributes, ordered attributes, symbolic categorical attributes, and character attributes. A hybrid calculation method using cosine similarity and Jaccard similarity coefficient can be used. Based on the similarity calculation results, algorithms such as nearest neighbor and K-nearest neighbor can be used to obtain retrieval results.

[0166] The above similarity calculation process is:

[0167] As shown in the above data structure, there are different methods for ordering, mainly including text description, logical classification and numerical representation. Among them:

[0168] The text description similarity d1 can be calculated using vocabulary similarity algorithms such as TF-IDF, Doc2Vec, and Word2Vec;

[0169] The logical classification similarity d2 can be calculated using the following algorithm formula:

[0170]

[0171] Where xi represents the current case, yi represents the historical case, and n represents the number of relevant information related to the case. The relevant information includes at least one of the following: station-related information, car order-related information, vehicle-related information, driver-related information, route-related information, and traffic signal information-related information. The similarity between the current case and the historical case can be calculated through the logical classification similarity d2;

[0172] The numerical representation of similarity d3 can be calculated using the following algorithm formula:

[0173]

[0174] ωi represents a predefined weight value, which is the weight value of the difference between the current case and the historical case, αi=xi max -xi min ,xi maxrepresents the maximum value of relevant information related to the case, xi min Represents the minimum value of relevant information related to the case, and this calculation process is used to achieve parameter normalization.

[0175] In this scheme, the similarity of approval cases refers to the similarity calculated for all attributes of the entire case. Therefore, the above algorithm can be used to calculate the similarity of each type separately and then perform weighted summation (each weight value can be assigned manually or automatically adjusted through manual evaluation). The specific formula is: dtotal = ω1×d1+ω2×d2+ω3×d3, dtotal represents the final calculated similarity, ω1 represents the weight of the calculation result obtained by text description similarity, ω2 represents the weight of the calculation result obtained by logical classification similarity, and ω3 represents the weight of the calculation result obtained by numerical representation similarity.

[0176] 3) Reuse and manual correction of approval results: Based on the retrieval results of similar cases based on the set model, if the new approval requirements are successfully matched, the case recommendation result with the highest similarity is output and recommended for automatic execution; otherwise, manual approval and correction steps are involved, and the corrected results are verified. The corrected results are also returned to the case reasoning system as high-quality samples for iterative learning by the intelligent model.

[0177] 4) Case Evaluation and Learning: With the continuous accumulation of post-approval cases and user feedback on approval results (i.e., approval case evaluation), this technical solution adopts the AutoML method to regularly perform self-adaptation and self-learning to maintain the approval sequence, so that the process approval reasoning results can be closer to business needs and logic, greatly improving the system's reasoning capabilities; approval cases can also be stored in the approval case library.

[0178] The above-mentioned processing module, selection module and first optimization module first initialize the approval order, and then optimize the approval order of different categories. This can continuously simplify and optimize the approval order of coal mines and avoid the complexity of the approval order.

[0179] For the coal mine transportation process, it includes site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency task allocation, etc. Therefore, the above links can be optimized. In another embodiment of the present application, the second optimization unit includes a second optimization module, and the second optimization module is used to optimize the site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency tasks of the transportation process based on the above transportation needs. In this embodiment, by optimizing the site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency tasks of the transportation process, efficient coordination and allocation of various resources in the auxiliary transportation process can be achieved.

[0180] Specifically, an AI resource scheduling optimization engine can be used to optimize the transportation process. The AI resource scheduling optimization engine is oriented towards the previously approved vehicle order requirements. Based on the collected historical / real-time data, it uses big data, artificial intelligence, operations optimization and other technologies to achieve intelligent perception and real-time global optimization of the core elements of the auxiliary transportation system, such as personnel, vehicles, materials, traffic lights, and driving routes.

[0181] In order to further optimize the transportation process more efficiently, optimization can be performed based on various relevant information of the transportation process. By obtaining multiple relevant information, the links of the transportation process can be optimized. In a specific embodiment of the present application, the second optimization module includes a third acquisition submodule, a first optimization submodule, a fourth acquisition submodule, a fifth acquisition submodule, a second optimization submodule, a sixth acquisition submodule and a third optimization submodule. The functions of each submodule are as follows:

[0182] A third acquisition submodule is configured to acquire site-related information, wherein the site-related information includes at least one of the following: coal mine roadway topology, miner information, and personnel operation distribution in the coal mine roadway;

[0183] The first optimization submodule is used to optimize the station and vehicle departure time based on the above-mentioned station-related information;

[0184] Specifically, the station configuration model can be used to optimize the station and vehicle departure time, such as Figure 5 As shown, the station configuration model is primarily designed for personnel transport vehicles with relatively fixed schedules, routes, and stations. It aims to reduce waiting times and walking distances during peak usage periods. The model uses a mine tunnel topology, miner information, feasible pre-set stations, and personnel work distribution as its primary inputs. Using a genetic algorithm, it calculates the locations of each station over a period of time, combines these stations into a route, and provides the starting departure time for the shuttle bus route.

[0185] A fourth acquisition submodule is configured to acquire information related to a vehicle order, wherein the information related to the vehicle order includes at least one of the following: a vehicle waiting time, a vehicle cost, and a vehicle allocation order;

[0186] A fifth acquisition submodule is configured to acquire vehicle-related information, wherein the vehicle-related information includes at least one of the following: vehicle type and load capacity;

[0187] A second optimization submodule is configured to optimize the vehicle dispatch process based on the vehicle order information and the vehicle information;

[0188] Specifically, the vehicle scheduling optimization model can be used to optimize the vehicle scheduling process, such as Figure 5 As shown in the figure, the vehicle scheduling optimization model, combined with vehicle demand priorities, uses vehicle demand (including planned and temporary use) and available vehicle information as primary inputs. It selects the optimal vehicle to perform specific transport tasks, targeting minimum vehicle empty mileage, vehicle waiting time, and vehicle cost. By splitting large demand orders into small transport work orders and then combining small demand orders into single orders, the model implements a logistics warehouse management model for vehicle scheduling that responds to multiple vehicles per order, or multiple orders per vehicle, thereby improving vehicle transport efficiency. The model's key constraints include accepting all high-priority orders, matching vehicle transport capacity with work order requirements, ensuring that different work order schedules do not conflict, and ensuring that the vehicle's mileage is no less than the transport distance.

[0189] A sixth acquisition submodule is configured to acquire driver-related information, wherein the driver-related information includes at least one of the following: accumulated mileage, accumulated working hours, driving qualifications, and number of dispatches;

[0190] The third optimization submodule is configured to optimize the driver's dispatching process based on the driver-related information.

[0191] Specifically, the driver scheduling model can be used to optimize the driver scheduling process, such as Figure 5 As shown, the driver scheduling model assigns drivers to each vehicle assigned to a work order in the vehicle scheduling optimization model based on demand order information and driver information (including accumulated mileage, accumulated working hours, driving qualifications, and number of dispatches). This model aims to balance each driver's recent accumulated mileage, time, and wages. It can be solved using operations optimization or heuristic algorithms, and primarily includes constraints such as matching driver and vehicle qualifications, requiring each vehicle to be driven by a human, and requiring each driver to drive a maximum of one vehicle.

[0192] The above-mentioned third acquisition submodule, first optimization submodule, fourth acquisition submodule, fifth acquisition submodule, second optimization submodule, sixth acquisition submodule and third optimization submodule can optimize the departure time of the station and the vehicle through the station-related information, optimize the vehicle scheduling process through the vehicle-related information, and optimize the driver scheduling process through the driver-related information, thereby further realizing the efficient coordination and allocation of various resources in the auxiliary transportation process.

[0193] In order to further optimize the transportation process more efficiently, optimization can be performed based on various relevant information of the transportation process. By obtaining multiple relevant information, the links of the transportation process can be optimized. In another specific embodiment of the present application, the second optimization module includes a seventh acquisition submodule, a fourth optimization submodule, an eighth acquisition submodule, a fifth optimization submodule and a processing submodule. The functions of each submodule are as follows:

[0194] a seventh acquisition submodule, configured to acquire path-related information, wherein the path-related information includes at least one of the following: road conditions of underground coal mine tunnels, geometric representations of the tunnels, map information of the underground coal mine tunnels, environmental parameter change information, and a vehicle's starting and ending points;

[0195] Specifically, the environmental parameter change information refers to information about changes in the environment of the mine, including at least information about changes in dust, information about changes in smoke, information about whether there is water in the mine, and information about whether there is danger in the mine.

[0196] A fourth optimization submodule is used to optimize the path planning process based on the above path related information;

[0197] Specifically, the path optimization model can be used to optimize the route of the vehicle to perform the transportation task, such as Figure 5 As shown in the figure, assisted transport drivers often face the problem of incorrect driving routes and unclear drivable routes after changes in working faces or tunnels. The intelligent path planning model provides recommended routes for drivers based on coal mine map information, transport work order information (including pick-up and drop-off points), vehicle attribute information, and real-time status monitoring data. It can also inform drivers of subsequent driving behaviors in advance based on the vehicle route and current status. When driving on the wrong road, it can prompt and re-plan the route. Compared with road traffic planning, the path planning model has the particularity of industry scenarios and is more complex to implement. Differentiated data processing and business modeling are required for this business. The model takes the minimum driving path and driving time as its main goals, and can be solved using a variety of technologies such as the A-Star algorithm, heuristic algorithm, and reinforcement learning.

[0198] an eighth acquisition submodule, configured to acquire information related to traffic signal information, wherein the information related to traffic signal information includes at least one of the following: vehicle distribution, vehicle speed, vehicle type, position sensing information corresponding to the vehicle, and road conditions of coal mine tunnels;

[0199] Specifically, the position sensing information corresponding to the vehicle refers to information sensed by a coil at a position corresponding to the vehicle.

[0200] A fifth optimization submodule, configured to optimize the configuration process of traffic signal information based on the above-mentioned traffic signal information related information;

[0201] Specifically, the traffic signal information setting model can be used to optimize the configuration process of traffic signal information, such as Figure 5As shown, this model achieves the following business objectives by collaboratively controlling underground traffic signal information: reducing unnecessary vehicle stops and waiting times and ensuring future driving safety; providing an alarm message and taking over traffic signal control at the intersection when the induction coil of the traffic signal information system at the intersection fails; and taking over traffic signal control at the intersection when a high-priority emergency task occurs to ensure smooth passage of emergency vehicles. This solution simulates and trains underground traffic signal scheduling using deep reinforcement learning methods. Multiple agents independently control each signal light, while introducing mutual information for collaboration, achieving real-time optimization and scheduling of underground traffic signal information settings.

[0202] The processing submodule is used to obtain the priority order of multiple dispatched tasks, and when an urgent task is received, set the priority of the urgent task to the highest priority.

[0203] Specifically, if Figure 5 As shown, an emergency mission model can be used to integrate the five aforementioned models: site configuration, vehicle scheduling, driver scheduling, route planning, and traffic signal information control. For emergency missions at the security level, comprehensive coordinated control and optimization are performed to ensure the highest priority for resource allocation and the timely arrival of required vehicles and personnel. This model requires high response time, integrating historical data to predict the status of the auxiliary transportation system in future time periods. Furthermore, heuristic algorithms or reinforcement learning algorithms are used for decision optimization to accelerate the solution process and achieve rapid response.

[0204] The above-mentioned seventh acquisition submodule, fourth optimization submodule, eighth acquisition submodule, fifth optimization submodule and processing submodule can optimize the path planning process through path-related information, optimize the traffic signal information configuration process through traffic signal information-related information, and can also coordinate other optimization processes in the event of an emergency task, thereby further realizing the efficient coordination and allocation of various resources in the auxiliary transportation process.

[0205] The above-mentioned global optimization device for auxiliary transportation of coal mines includes a processor and a memory. The above-mentioned acquisition unit, first optimization unit and second optimization unit, etc. are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0206] The processor contains a core, which retrieves the corresponding program unit from the memory. One or more cores can be set, and the efficiency of coal mine transportation can be improved by adjusting the core parameters.

[0207] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0208] An embodiment of the present invention provides a computer-readable storage medium having a program stored thereon, which implements the above-mentioned global optimization method for auxiliary transportation of coal mines when the program is executed by a processor.

[0209] An embodiment of the present invention provides a processor, which is used to run a program, wherein the global optimization method for auxiliary transportation of coal mines is executed when the program is run.

[0210] The present application also provides an electronic device, comprising one or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include a method for executing any one of the above methods.

[0211] An embodiment of the present invention provides a device, comprising a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, at least the following steps are performed:

[0212] Step S101: Acquire approval process information and transportation requirements. The approval process information is information related to the approval order of work orders in the transportation process. The transportation requirements refer to transportation equipment requirements or transportation personnel requirements.

[0213] Step S102: Optimize the approval sequence of the transportation process based on the above approval process information to obtain an optimized approval sequence;

[0214] Step S103: Approve the transportation demand in accordance with the optimized approval order to obtain an approval result. If the approval result indicates approval, optimize the transportation process according to the transportation demand to obtain an optimized transportation process.

[0215] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0216] The present application also provides a computer program product, which, when executed on a data processing device, is adapted to execute a program for initializing at least the following method steps:

[0217] Step S101: Acquire approval process information and transportation requirements. The approval process information is information related to the approval order of work orders in the transportation process. The transportation requirements refer to transportation equipment requirements or transportation personnel requirements.

[0218] Step S102: Optimize the approval sequence of the transportation process based on the above approval process information to obtain an optimized approval sequence;

[0219] Step S103: Approve the transportation demand in accordance with the optimized approval order to obtain an approval result. If the approval result indicates approval, optimize the transportation process according to the transportation demand to obtain an optimized transportation process.

[0220] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0221] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the above-mentioned units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0222] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0223] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0224] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the above-mentioned methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0225] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0226] 1) The global optimization method for auxiliary transportation of coal mines of this application first obtains approval process information and transportation demand, then optimizes the approval order of the transportation process according to the approval process information to obtain the optimized approval order, and finally approves the transportation demand according to the optimized approval order to obtain the approval result, and when the approval result indicates that the approval is passed, optimizes the transportation process according to the transportation demand to obtain the optimized transportation process. In this scheme, the approval order is continuously optimized, which can simplify the approval process, and the approval can be carried out in an intelligent manner. The resource scheduling of the transportation process can also be optimized, that is, the entire process of auxiliary transportation of coal mines is globally optimized, thereby improving the efficiency of auxiliary transportation of mines.

[0227] 2) The global optimization device for auxiliary transportation of coal mines of the present application, the acquisition unit acquires the approval process information and the transportation demand, the first optimization unit optimizes the approval order of the transportation process according to the approval process information, and obtains the optimized approval order, the second optimization unit approves the transportation demand according to the optimized approval order, obtains the approval result, and when the approval result indicates that the approval is passed, the transportation process is optimized according to the transportation demand, and the optimized transportation process is obtained. In this solution, the approval order is continuously optimized, which can simplify the approval process, and the approval can be performed in an intelligent manner. The resource scheduling of the transportation process can also be optimized, that is, the entire process of auxiliary transportation of coal mines is globally optimized, thereby improving the efficiency of auxiliary transportation of mines.

[0228] The above description is merely a preferred embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A global optimization method for auxiliary transportation in coal mines, characterized in that: include: Acquire approval process information and transportation requirements. The approval process information is information related to the approval order of work orders in the transportation process. The transportation requirements refer to transportation equipment requirements or transportation personnel requirements. Optimizing the approval sequence of the transportation process according to the approval process information to obtain an optimized approval sequence; Approving the transportation demand according to the optimized approval order to obtain an approval result, and if the approval result indicates approval, optimizing the transportation process according to the transportation demand to obtain an optimized transportation process; There are multiple approval sequences, and according to the approval process information, the approval sequence of the transportation process is optimized to obtain the optimized approval sequence, including: initializing the approval sequence, classifying the approval sequence according to the approval process information to obtain the classified approval sequence; selecting multiple approval sequences corresponding to the target category according to the transportation demand; selecting one approval sequence from the multiple approval sequences as the target approval sequence, and optimizing the approval nodes of the target approval sequence using a genetic algorithm to obtain the optimized target approval sequence, wherein the approval sequence includes at least one approval node, and the process of optimizing the approval nodes of the target approval sequence using a genetic algorithm includes: initializing the population, constructing an agent code, wherein the population represents the approval sequence, and the agent code represents the approval department and approver in the approval process; establishing a fitness function, and the fitness function is used to Make the approval process pass through the least number of approval departments, approval nodes and approvers, and at the same time make the approval process have the maximum safety value, the safety value represents the credit evaluation value of the vehicle demander; perform selection operation, the selection operation is used to determine the reorganization or crossover of individuals in the population, and determine the number of offspring individuals that the selected individuals will produce; perform crossover and mutation operations, the crossover and mutation operations include crossover operations and mutation operations in the coal mine auxiliary transportation scenario, the crossover operation is to exchange the approval nodes in the gene sequence to produce a new sub-agent, and the mutation operation is divided into replacing the current approval department with another approval department and replacing the current approver with another approver; determine whether the termination condition is met, and if the termination condition is met, end the optimization process of the approval order, and if the termination condition is not met, re-establish the fitness function, and re-execute the selection operation and the crossover and mutation operations in sequence, The transportation process is optimized according to the transportation demand to obtain an optimized transportation process, including: optimizing site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency tasks of the transportation process based on the transportation demand.

2. The method according to claim 1, characterized in that Optimizing the approval nodes of the target approval sequence to obtain an optimized target approval sequence includes: Acquire first approval information corresponding to the approval node, where the first approval information at least includes historical approval records; According to the first approval information, some approval nodes in the target approval sequence are deleted to obtain an optimized target approval sequence.

3. The method according to claim 1, characterized in that Optimizing the approval nodes of the target approval sequence to obtain an optimized target approval sequence includes: Obtaining second approval information corresponding to the approval node, the second approval information at least including user feedback data, historical approval records, and approval time; According to the second approval information, some approval nodes in the target approval sequence are replaced to obtain an optimized target approval sequence.

4. The method according to claim 1, wherein Optimizes site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration, and emergency tasks during the transportation process, including: Acquiring site-related information, wherein the site-related information includes at least one of the following: coal mine roadway topology, miner personnel information, and personnel operation distribution in the coal mine roadway; Optimize the station and vehicle departure time based on the station-related information; Obtaining information related to a vehicle order, wherein the information related to the vehicle order includes at least one of the following: a vehicle waiting time, a vehicle cost, and a vehicle allocation order; Acquire vehicle-related information, wherein the vehicle-related information includes at least one of the following: vehicle type and load capacity; Optimizing the vehicle dispatch process based on the vehicle order information and the vehicle information; Obtaining driver-related information, the driver-related information including at least one of the following: accumulated mileage, accumulated working hours, driving qualifications, and number of dispatches; The driver's dispatch process is optimized according to the driver-related information.

5. The method according to claim 1, wherein Optimizes site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration, and emergency tasks during the transportation process, including: Acquiring path-related information, the path-related information including at least one of the following: road conditions of underground coal mine tunnels, tunnel geometry, map information of underground coal mine tunnels, environmental parameter change information, and a vehicle's starting point and ending point; Optimizing the path planning process based on the path-related information; Obtaining information related to traffic signal information, wherein the information related to traffic signal information includes at least one of the following: vehicle distribution, vehicle speed, vehicle type, position sensing information corresponding to the vehicle, and road conditions of coal mine tunnels; Optimizing the configuration process of traffic signal information based on the traffic signal information related information; Get the priority order of multiple dispatched tasks, and when an urgent task is received, set the priority of the urgent task to the highest priority.

6. A global optimization device for auxiliary transportation of coal mines, characterized in that: include: An acquisition unit, configured to acquire approval process information and transportation requirements, wherein the approval process information is information related to the approval order of work orders in the transportation process, and the transportation requirements are transportation equipment requirements or transportation personnel requirements; A first optimization unit is configured to optimize the approval sequence of the transportation process according to the approval process information to obtain an optimized approval sequence; The second optimization unit is used to approve the transportation demand according to the optimized approval order to obtain an approval result, and when the approval result indicates that the approval is passed, optimize the transportation process according to the transportation demand to obtain an optimized transportation process. There are multiple approval sequences, and the first optimization unit includes a processing module, a selection module, and a first optimization module. The processing module is used to initialize the approval sequence, classify the approval sequence according to the approval process information, and obtain a classified approval sequence; The selection module is used to select multiple approval orders corresponding to the target category according to the transportation demand; The first optimization module is used to select an approval sequence from multiple approval sequences as a target approval sequence, and optimize the approval nodes of the target approval sequence to obtain an optimized target approval sequence, wherein the approval sequence includes at least one approval node, wherein the process of optimizing the approval nodes of the target approval sequence using a genetic algorithm includes: initializing a population, constructing an agent code, wherein the population represents the approval sequence, and the agent code represents the approval department and approver in the approval process; establishing a fitness function, wherein the fitness function is used to make the approval process pass through the least number of approval departments, approval nodes, and approvers, and at the same time make the approval process have the maximum safety value, wherein the safety value represents the safety of the vehicle. The credit rating value of the demander; performing a selection operation, the selection operation is used to determine the individuals in the population to be reorganized or crossed, and to determine the number of offspring individuals that the selected individuals will produce; performing a crossover and mutation operation, the crossover and mutation operation includes a crossover operation and a mutation operation in the coal mine auxiliary transportation scenario, the crossover operation is to exchange the approval nodes in the gene sequence to generate a new sub-agent, and the mutation operation is divided into replacing the current approval department with another approval department and replacing the current approver with another approver; determining whether the termination condition is met, and if the termination condition is met, ending the optimization process of the approval order; if the termination condition is not met, re-establishing the fitness function, and re-executing the selection operation and the crossover and mutation operation in sequence, The second optimization unit includes a second optimization module, which is used to optimize the site configuration, vehicle scheduling, driver scheduling, route planning, traffic signal information configuration and emergency tasks in the transportation process based on the above-mentioned transportation needs.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 5.

8. An electronic device, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for executing the method according to any one of claims 1 to 5.

Images