A coal storage and railway distribution collaborative intelligent logistics management and control system

Abstract

The application discloses a coal storage and railway distribution collaborative intelligent logistics management and control system, and belongs to the technical field of intelligent logistics and supply chain management and control. The system comprises a collaborative intelligent logistics management and control platform, and an intelligent storage management subsystem, a railway transportation scheduling subsystem, a data middle station, an Internet of Things sensing layer and an intelligent management and control center which are respectively in communication connection with the platform; the collaborative intelligent logistics management and control platform is internally provided with a collaborative decision engine and a closed-loop feedback module, the collaborative decision engine comprises three modules of demand prediction, supply-demand matching and optimized scheduling, and can complete collaborative optimized decision of storage inventory and railway transport capacity based on multi-source heterogeneous data. The application breaks through the business data island by unifying the data middle station, realizes real-time sensing of the whole process of storage, loading and railway in-transit by relying on the Internet of Things, effectively solves the problems of split information of traditional coal logistics storage and railway scheduling, supply-demand imbalance and transport capacity waste, improves the overall logistics efficiency, and reduces the comprehensive operation cost.

Description

Technical Field

[0001] This invention relates to the field of smart logistics and supply chain management technology, and in particular to a collaborative smart logistics management system for coal storage and railway distribution. Background Technology

[0002] Coal, as my country's primary energy source, involves a complex supply chain process in its logistics, encompassing coal mine production, warehousing, and railway shipment. In traditional coal logistics management models, warehousing management and railway transport scheduling are often handled by different departments using independent information systems, creating significant information silos. On the one hand, warehousing cannot access real-time railway capacity information, leading to both inventory backlogs and empty warehouses awaiting supplies. On the other hand, railway dispatching lacks real-time awareness of warehousing inventory status, often resulting in a mismatch between capacity allocation and actual demand.

[0003] In addition, the existing coal logistics management and control system has the following technical defects: 1. Lack of collaborative optimization mechanism: Warehouse management and railway transportation operate independently, lacking a unified collaborative decision-making platform. This makes it impossible to achieve real-time matching and dynamic optimization of warehouse inventory status and railway transportation capacity resources, resulting in low overall logistics efficiency and high operating costs.

[0004] 2. Lack of intelligent forecasting and early warning capabilities: Most existing systems rely on human experience and simple statistical reports for decision-making, and cannot make intelligent forecasts based on multi-source information such as historical data, market demand and weather conditions, making it difficult to detect the risk of supply and demand imbalance in a timely manner and issue early warnings.

[0005] 3. Severe data silos and lack of a unified data foundation: Warehousing data, transportation data, and order data are scattered across different business systems, lacking unified data collection, cleaning, integration, and analysis capabilities, and thus failing to form data assets that support intelligent decision-making.

[0006] 4. Outdated management and control methods and lack of closed-loop feedback: The existing management and control model is mainly based on post-event statistics and lacks real-time monitoring, abnormal early warning and closed-loop feedback optimization mechanism for the entire logistics process, making it difficult to achieve complete closed-loop management from planning and execution to feedback.

[0007] Therefore, there is an urgent need for a smart logistics management and control system that can deeply integrate coal storage management with railway distribution and scheduling, and achieve end-to-end data connectivity and intelligent collaborative optimization. Summary of the Invention

[0008] (1) Technical problems to be solved 1. Lack of collaborative optimization mechanism between warehousing and railway distribution: In traditional coal logistics systems, warehousing management and railway transportation scheduling are separate systems with no information sharing, resulting in both inventory backlog and transportation capacity waste, and making it impossible to dynamically optimize scheduling based on real-time inventory status and transportation resources.

[0009] 2. Lack of intelligent forecasting and early warning capabilities based on multi-source data: Existing systems cannot effectively integrate multi-source data such as warehousing, transportation, market, and weather, and lack the ability to forecast supply and demand, provide inventory warnings, and match transportation capacity based on big data and intelligent algorithms.

[0010] 3. Severe data silos and lack of a unified data foundation: Warehouse data, railway scheduling data, customer order data, and IoT sensing data are scattered in different systems, lacking a unified data collection, cleaning, integration, and analysis platform.

[0011] 4. Lack of a closed-loop management and feedback optimization mechanism for the entire process: The existing management and control model is mainly based on manual experience and post-event statistics, which cannot achieve real-time monitoring, anomaly warning and adaptive optimization of the entire process of warehousing, loading, shipping and in transit.

[0012] (2) Technical solution To achieve the above objectives, the present invention provides a collaborative intelligent logistics management and control system for coal storage and railway distribution, including a collaborative intelligent logistics management and control platform, and an intelligent storage management subsystem, a railway transportation dispatching subsystem, a data platform, an Internet of Things sensing layer, and an intelligent management and control center, which are respectively connected to the collaborative intelligent logistics management and control platform. The system adds a closed-loop feedback module and connects with the customer order management system and the railway dispatching center through a standardized API interface to achieve real-time interaction of orders, transportation capacity, and dispatching instructions.

[0013] The collaborative intelligent logistics management and control platform is the core hub of the entire system, with a built-in collaborative decision engine and closed-loop feedback module. The collaborative decision engine includes a demand forecasting module, a supply and demand matching module, and an optimization scheduling module, which are used to make collaborative optimization decisions on warehouse inventory and railway capacity based on multi-source data. The closed-loop feedback module collects actual execution data on site, compares it with planned data to calculate deviations, and sends it back to the collaborative decision engine to complete the adaptive optimization of the scheduling plan.

[0014] The intelligent warehouse management subsystem includes an inbound management module, an inventory monitoring module, and an outbound scheduling module, which are used for digital management of the entire coal storage process. The inventory monitoring module collects real-time data on the inventory quantity, quality indicators, storage time, and coal pile temperature of each coal type and each stack, and synchronizes it to the data platform. When the coal pile temperature is detected to exceed the safety threshold, a priority outbound instruction is automatically issued.

[0015] The railway transportation scheduling subsystem includes a train number planning module, a carriage loading module, and a dispatch scheduling module, which are used for digital management of the entire railway dispatch process. The train number planning module generates train number plans based on the dispatch instructions output by the collaborative decision engine and the railway timetable. The carriage loading module performs intelligent loading based on coal type, particle size, density, and carriage load limits to avoid overloading and uneven loading.

[0016] The data platform includes a data acquisition layer, a data cleaning layer, a data storage layer, and a data analysis layer. It is used to uniformly collect, clean, store, and analyze multi-source heterogeneous data to form data assets that support intelligent decision-making. The data acquisition layer has local caching and breakpoint resume functions. When the external interface is disconnected, the data is temporarily stored and automatically retransmitted after the link is restored to prevent data loss.

[0017] The IoT sensing layer includes sensors, RFID readers, video surveillance equipment, weighbridges / railway scales, and GPS / BeiDou positioning terminals deployed at storage stations and along railway lines. These are used to sense the real-time status of coal storage and transportation, and to upload the sensed data to the data platform.

[0018] The intelligent control center includes a visualization screen module, an anomaly warning module, and an intelligent decision-making module, which are used to perform visual monitoring, anomaly warning, and intelligent decision support for the entire process of coal storage and railway distribution.

[0019] Furthermore, the demand forecasting module in the collaborative decision-making engine is based on ARIMA, Prophet time series model, XGBoost, and LSTM machine learning algorithms, and overlays external influencing factors such as market demand, meteorological conditions, seasonal coal consumption, and railway maintenance to forecast coal demand, inventory consumption, and railway transport capacity demand at multiple scales: short-term (1-7 days), medium-term (1-4 weeks), and long-term (1-12 months).

[0020] Furthermore, the supply and demand matching module in the collaborative decision engine is based on the NSGA-II and MOEA / D multi-objective optimization algorithms to establish a matching model between warehouse inventory, customer demand and railway transport capacity. The objective functions are to minimize inventory costs, maximize transportation efficiency and optimize customer satisfaction. The module also incorporates constraints on warehouse capacity, train load, delivery time window and safe storage period of coal piles to achieve a multi-objective balance.

[0021] Furthermore, the optimization scheduling module in the collaborative decision engine uses genetic algorithms, ant colony algorithms, or particle swarm optimization algorithms to perform joint global optimization of the warehouse outbound sequence, loading plan, and vehicle schedule, generating a globally optimal collaborative scheduling scheme.

[0022] Furthermore, the data analysis layer of the data platform includes an inventory analysis module, a transportation capacity analysis module, a demand analysis module, and a cost analysis module, and uses multidimensional data cube OLAP and graph database technology to perform multidimensional correlation analysis on the data.

[0023] Furthermore, the sensors in the IoT sensing layer include temperature and humidity sensors, coal pile temperature sensors, dust concentration sensors, and material level sensors, which are used to monitor the safety of the storage environment and the inventory status of the stacks 24 hours a day in real time.

[0024] Furthermore, the anomaly early warning module of the intelligent control center, based on preset thresholds and the Isolated Forest and LOF intelligent anomaly detection algorithms, provides real-time early warnings for anomalies such as over- or under-stocked inventory, overloaded / uneven loading, train delays / route deviations, excessive coal temperature in the storage yard, and excessive dust. Alarm information is pushed through multiple channels, including large-screen pop-ups, WeChat, SMS, and mobile apps.

[0025] (3) Beneficial effects In summary, the present invention has the following beneficial effects: 1. Deep collaboration between warehousing and railway distribution has been achieved: By building a collaborative intelligent logistics management and control platform, the intelligent warehousing management subsystem and the railway transportation scheduling subsystem are deeply integrated. Based on the collaborative decision engine, the real-time matching and dynamic optimization of warehousing inventory status and railway transportation capacity resources are realized, which effectively solves the problem of coexistence of inventory backlog and transportation capacity waste, and significantly improves the overall logistics efficiency.

[0026] 2. Enhanced intelligent forecasting and early warning capabilities: By integrating multi-source heterogeneous data through a data platform, and utilizing intelligent algorithms such as machine learning and time series analysis, demand, inventory, and transportation capacity are accurately predicted. Based on the forecast results, early warnings and proactive scheduling are carried out, significantly reducing the risk of supply and demand imbalance.

[0027] 3. Data silos have been broken down and a unified data foundation has been built: Through the data platform, warehousing data, transportation data, order data and IoT sensing data are collected, cleaned, stored and analyzed in a unified manner, forming data assets that support intelligent decision-making and providing a solid data foundation for full-process digital management and control.

[0028] 4. Achieved closed-loop control and adaptive optimization throughout the entire process: Real-time perception of the entire process is realized through the Internet of Things perception layer, visual monitoring and anomaly warning are realized through the intelligent control center, and comparative analysis and adaptive optimization of planning and execution are realized through the closed-loop feedback module, forming a complete closed-loop management system from planning, execution to feedback.

[0029] 5. Reduced operating costs and improved resource utilization: Through collaborative and optimized scheduling, inventory backlog and empty warehouse waiting were reduced, railway capacity utilization was improved, and logistics operating costs were reduced. At the same time, accurate forecasting and proactive scheduling improved the timeliness and accuracy of order delivery, thereby improving overall operational efficiency. Attached Figure Description

[0030] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only one embodiment of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 This is a diagram of the overall system architecture of the present invention; Figure 2 This is a flowchart illustrating the collaborative business process of coal storage and railway distribution in this invention. Figure 3 This is the intelligent decision-making and data flow diagram of the present invention. Detailed Implementation

[0032] To make the technical means, creative features, objectives and effects of this invention easier to understand, the technical solutions in the specific embodiments of this invention are described clearly and completely below to further illustrate this invention. Obviously, the specific embodiments described are only a part of the embodiments of this invention, and not all of them.

[0033] like Figure 1 As shown, the intelligent logistics management and control system for coal storage and railway distribution described in this invention adopts a layered architecture design. Externally, it interfaces with the customer order management system and the railway dispatch center. The bottom layer is the IoT sensing layer, responsible for collecting raw data from the storage yard, loading, and train sites and uploading it to the data platform. After unified data governance, the data platform inputs it into the core layer of the intelligent logistics management and control platform. The platform has a built-in collaborative decision-making engine and a closed-loop feedback module, which issue execution instructions to the intelligent storage management subsystem and the railway transportation dispatch subsystem, respectively. The top layer is the intelligent control center, which realizes full-process visual monitoring and alarms. The closed-loop feedback module connects the execution layer and the decision engine, forming a continuously iterative optimization loop.

[0034] The demand forecasting module of the collaborative decision-making engine adopts an improved LSTM neural network, adding industry-specific parameters such as safe storage time of coal piles, transportation disruptions due to rain and snow, and monthly railway maintenance plans to the input features. Compared with general forecasting models, the accuracy of medium- and long-term demand forecasting is significantly improved. The supply and demand matching module adopts an improved NSGA-II algorithm, adding a hard constraint of prioritizing the outbound shipment from high-temperature stacks, balancing transportation efficiency and fire safety in the storage yard. The optimization scheduling module optimizes loading sequence based on particle swarm optimization algorithm, reducing cross-congestion and waiting time of transfer vehicles in the yard.

[0035] The data platform uses InfluxDB to store time-series sensor data, PostgreSQL to store structured business data such as orders and train numbers, and Neo4j to build a relationship graph of stacking positions, orders, and trains. The data acquisition layer is configured with disconnection caching and breakpoint resume mechanisms. When the external interface is interrupted, business data is temporarily stored locally and automatically retransmitted after the link is restored to ensure data integrity.

[0036] The IoT sensing layer independently deploys coal pile temperature sensors at each coal storage location, and the data is synchronized in real time to the abnormal early warning module of the smart control center. Once the coal pile temperature exceeds the safety threshold, the system automatically generates an emergency outbound dispatch instruction, pushes alarm information to the site management personnel, and links the outbound dispatch module to prioritize the outbound of coal from that location, thereby reducing the risk of spontaneous combustion.

[0037] The intelligent warehouse management subsystem's inbound management module can automatically identify incoming coal vehicles, collect weights through a track scale, synchronize coal quality test data, and allocate stacking locations according to indicators such as calorific value, sulfur content, and particle size. The outbound scheduling module receives outbound plans generated by the collaborative decision engine, optimizes the outbound sequence and loading path by combining the spatial distribution of stacking locations and transfer paths, and improves outbound operation efficiency.

[0038] The train number planning module of the railway transportation dispatching subsystem combines the railway bureau's operation map with available carriage resources to compile formal train number plans; the carriage loading module intelligently allocates loading volume according to the physical characteristics of coal, carriage type and load limit to ensure uniform loading and prevent overloading and uneven loading; the dispatching module tracks the entire process of loading, marshalling and departure in real time and synchronizes the train number status to the railway dispatching center.

[0039] The intelligent control center's visualization screen module integrates GIS maps, 3D warehouse stack models, and dynamic statistical charts to provide a panoramic view of inventory status, loading progress, train locations en route, and order delivery status. The intelligent decision-making module supports scheduling scheme simulation, supply and demand scenario deduction, and one-click issuance of the optimal solution to assist managers in making decision-making interventions.

[0040] like Figure 2 As shown, the collaborative business process of the present invention includes the following steps: S1: Customer Order Receipt. The system receives order information from the customer order management system via an API interface, including coal type, quantity, delivery time, and delivery location.

[0041] S2: Demand Forecasting and Planning. The demand forecasting module of the collaborative decision engine uses order data, historical sales data, and external influencing factors to make multi-scale forecasts of demand for future periods, generating preliminary warehouse outbound plans and railway shipment plans.

[0042] S3: Inventory Check. The inventory monitoring module retrieves the inventory status of each type of coal in real time to determine whether the inventory meets demand.

[0043] S4: Inventory Replenishment Assessment. If inventory is insufficient, the coal replenishment process is triggered, coordinating with upstream coal mines or ports to supplement supply; if inventory is sufficient, proceed to the next step.

[0044] S5: Capacity Matching and Scheduling. The supply and demand matching module combines inventory status, customer demand, and available railway capacity to perform supply and demand matching and capacity allocation, generating a preliminary scheduling plan.

[0045] S6: Capacity Verification. If available capacity cannot meet demand, apply to the railway dispatch center for temporary capacity, such as adding train services or coordinating spare carriages; if capacity matches, generate a loading plan.

[0046] S7: Loading Plan Generation. The optimized scheduling module generates a detailed loading plan based on the supply and demand matching results, including the order of departure from the warehouse, the wagon loading scheme, and the loading time window.

[0047] S8: Shipment Execution. Loading operations are carried out according to the loading plan, and after completion, the railway transportation dispatching subsystem coordinates with the railway dispatching center to organize the shipment.

[0048] S9: On-the-Route Monitoring and Arrival Confirmation. Trains are tracked in real time via the IoT sensing layer, and delivery confirmation is completed upon arrival of goods.

[0049] S10: Closed-loop feedback optimization. The closed-loop feedback module collects actual execution data, compares and analyzes it with planned data, calculates the deviation, and sends it back to the collaborative decision engine to adjust the prediction model parameters and optimize the scheduling strategy, thereby achieving system adaptive optimization.

[0050] like Figure 3As shown, the intelligent decision-making data flow of this invention is as follows: multi-source data acquisition (warehousing, railway, orders, meteorology, IoT sensing) → data cleaning and preprocessing (ETL) → multi-type database storage (time series database, relational database, graph database) → multi-dimensional data analysis and intelligent prediction → collaborative decision engine computation (multi-objective optimization, supply and demand matching, dynamic scheduling) → issuance of warehousing scheduling instructions and railway scheduling instructions → on-site execution → execution data feedback → closed-loop optimization. During execution, actual outbound volume, actual loading time, actual arrival time, and other execution data are compared with planned data, and deviation information is fed back to the collaborative decision engine for iterative optimization of the prediction model and scheduling strategy.

[0051] This invention is not limited to the specific embodiments described above. Without departing from the spirit and essence of this invention, those skilled in the art can make various corresponding modifications and variations based on this invention, and all such modifications and variations should fall within the protection scope of the appended claims.

Claims

1. A collaborative intelligent logistics management and control system for coal storage and railway distribution, characterized in that, It includes a collaborative intelligent logistics management and control platform, and an intelligent warehouse management subsystem, a railway transportation dispatching subsystem, a data platform, an IoT sensing layer, and an intelligent management and control center, all of which are communicatively connected to the collaborative intelligent logistics management and control platform; the system connects to the customer order management system and the railway dispatching center through standardized API interfaces; The collaborative intelligent logistics management and control platform has a built-in collaborative decision engine and a closed-loop feedback module. The collaborative decision engine includes a demand forecasting module, a supply and demand matching module, and an optimization scheduling module, which are used to complete collaborative optimization decisions on warehousing inventory and railway transport capacity based on multi-source data. The closed-loop feedback module collects on-site execution data, compares it with planned data to obtain deviations, and sends them back to the collaborative decision engine to realize adaptive optimization of the scheduling scheme. The intelligent warehouse management subsystem includes an inbound management module, an inventory monitoring module, and an outbound scheduling module, which are used for digital control of the entire coal storage process. The inventory monitoring module collects real-time data on the inventory, coal quality, storage time, and coal temperature of each type of coal and synchronizes it to the data platform. When the coal temperature exceeds the standard, it automatically issues a priority outbound instruction. The railway transportation scheduling subsystem includes a train number planning module, a carriage loading module, and a departure scheduling module, which are used for digital management and control of the entire railway dispatch process. The train number planning module receives scheduling instructions from the collaborative decision engine to generate train plans, and the carriage loading module achieves intelligent loading by combining coal type characteristics and carriage load. The data platform comprises a data acquisition layer, a data cleaning layer, a data storage layer, and a data analysis layer, used for unified acquisition, cleaning, storage, and multidimensional analysis of multi-source heterogeneous data; the data acquisition layer has local caching and breakpoint resume functions; The IoT sensing layer includes various sensors, RFID readers, video surveillance, weighbridges / rail scales, and GPS / BeiDou positioning terminals deployed in warehousing stations and along railway lines, which collect real-time data on the status of warehousing and transportation sites and upload it to the data platform. The intelligent control center includes a visualization screen module, an anomaly warning module, and an intelligent decision-making module, which are used for full-process visual monitoring, anomaly alarms, and scheduling decision support.

2. The intelligent logistics management and control system for collaborative coal storage and railway distribution according to claim 1, characterized in that, The demand forecasting module integrates time series analysis models and machine learning algorithms, and overlays external industry characteristics such as market, weather, season, and railway maintenance to forecast coal demand, inventory consumption, and railway transport capacity demand at multiple scales, including short-term, medium-term, and long-term.

3. The intelligent logistics management and control system for collaborative coal storage and railway distribution according to claim 1, characterized in that, The supply and demand matching module uses a multi-objective optimization algorithm to construct a matching model for inventory, customer orders, and railway capacity. The goal is to minimize inventory costs, maximize transportation efficiency, and optimize customer delivery satisfaction, while taking into account constraints such as storage capacity, train load, delivery time, and safe storage period of coal piles to achieve a balance of multiple objectives.

4. The intelligent logistics management and control system for collaborative coal storage and railway distribution according to claim 1, characterized in that, The optimization scheduling module uses genetic algorithms, ant colony algorithms, or particle swarm optimization algorithms to perform joint global optimization of the order of warehouse exit, loading plan, and train number, and outputs the optimal collaborative scheduling scheme.

5. The intelligent logistics management and control system for collaborative coal storage and railway distribution according to claim 1, characterized in that, The data analysis layer includes sub-modules for inventory analysis, transportation capacity analysis, demand analysis, and cost analysis, and uses a multi-dimensional data cube and graph database to complete multi-dimensional correlation data analysis.

6. The intelligent logistics management and control system for collaborative coal storage and railway distribution according to claim 1, characterized in that, The IoT sensing layer sensors include material level sensors, coal pile temperature sensors, temperature and humidity sensors, and dust concentration sensors, which monitor the stockpile inventory status and environmental safety indicators in real time.

7. The intelligent logistics management and control system for collaborative coal storage and railway distribution according to claim 1, characterized in that, The anomaly warning module combines preset thresholds with the isolated forest intelligent anomaly detection algorithm to issue real-time alarms for anomalies such as over- or under-stocking of inventory, overloading or uneven loading of trucks, train delays, excessively high temperature of coal piles, and excessive dust. Alarm information is pushed through multiple channels, including large screen pop-ups, WeChat, SMS, and mobile APP.

8. A method for coordinated and optimized scheduling of coal storage and railway distribution, characterized in that, The control system applied to any one of claims 1-7 includes the following steps: S1: The system automatically receives order information from the customer order management system through a standardized API interface, including coal type, demand quantity, delivery time limit, and delivery location; S2: The collaborative decision-making engine demand forecasting module integrates multi-source data such as historical orders, market demand, weather, and railway maintenance to make multi-scale coal demand and transportation capacity demand forecasts and generate preliminary outbound and shipping plan baselines; S3: The intelligent warehouse management subsystem's inventory monitoring module retrieves real-time pallet inventory data to verify whether the existing inventory meets the predicted shipment demand. S4: If the inventory does not meet the demand, initiate the upstream coal replenishment process; if the inventory is sufficient, proceed to the transportation capacity matching stage. S5: The supply and demand matching module uses a multi-objective optimization algorithm based on real-time inventory, customer orders, and available railway capacity to complete the matching and allocation of inventory, demand, and capacity. S6: Verify whether the existing total railway transport capacity meets the shipping demand. If there is a capacity shortage, apply to the railway dispatch center for temporary additional wagons or spare wagons; if the capacity is sufficient, generate a loading plan. S7: The optimized scheduling module uses intelligent optimization algorithms to jointly optimize the order of warehouse exit, carriage loading scheme and train scheduling, and outputs the globally optimal collaborative scheduling scheme. S8: The collaborative intelligent logistics management and control platform breaks down the scheduling plan into warehouse outbound execution instructions and railway dispatch instructions, which are then sent to the intelligent warehouse management subsystem and the railway transportation scheduling subsystem for execution, respectively. S9: The IoT sensing layer collects real-time data on loading weighing, train positioning, and yard environment perception, and uploads it to the data platform. The smart control center monitors the entire process in real time and automatically identifies abnormal alarms. S10: The closed-loop feedback module summarizes all actual execution data on site, compares it with the planned scheduling data to calculate the deviation, and sends the deviation dataset back to the collaborative decision engine. The engine updates the prediction model parameters and optimization algorithm constraints based on online learning to achieve adaptive iterative optimization of the scheduling scheme.

9. The collaborative optimization scheduling method according to claim 8, characterized in that, The supply and demand matching module adopts the NSGA-II or MOEA / D multi-objective optimization algorithm, with the objective functions of minimizing inventory cost, maximizing transportation efficiency, and optimizing customer delivery satisfaction, and with the constraints of storage capacity, train load, customer delivery time window, and safe storage period of coal pile as constraints, and outputs multiple sets of Pareto optimal supply and demand matching schemes.

Images