Intelligent unmanned warehouse scheduling system for spare parts

Abstract

The invention provides an intelligent unmanned library scheduling system for spare parts. The intelligent unmanned library scheduling system comprises an information perception module, a demand prediction analysis module, an intelligent scheduling core and an unmanned library execution monitoring module. The system obtains spare part information in all directions, predicts future demands, generates scheduling strategies, controls unmanned equipment to complete carrying and storage operations, monitors equipment states and environmental parameters, and realizes intelligent scheduling of multi-objective optimization. According to the invention, the efficiency and accuracy of spare part management can be improved, the inventory cost is reduced, the efficient operation of equipment and the stability of environmental parameters are ensured, and the response speed and service quality of the whole supply chain are optimized.

Description

Technical Field

[0001] The present invention relates to the field of intelligent logistics, and more specifically, to an intelligent unmanned warehouse scheduling system for spare parts. Background Art

[0002] In modern industrial production and logistics management, effective spare parts management is crucial for ensuring continuous production line operation and reducing maintenance costs. Traditional spare parts management relies on manual operations and simple inventory systems, which often lack real-time monitoring and accurate demand forecasting capabilities, leading to inventory backlogs or spare parts shortages, which in turn impact production efficiency and the timeliness of equipment maintenance. Furthermore, manual management methods are inefficient, error-prone, and difficult to adapt to rapidly changing production demands and equipment status. With the development of the Internet of Things, big data, and artificial intelligence technologies, intelligent warehouse management systems have gradually become an effective means of improving logistics efficiency and reducing costs. These systems can achieve precise tracking, storage, and scheduling of spare parts through automation, but many existing systems still lack multi-objective optimization, real-time monitoring, and environmental adaptability.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: the existing spare parts management system is often unable to fully obtain detailed information on spare parts, demand forecasts are not accurate enough, scheduling strategies are single and lack flexibility, and the monitoring system does not respond to environmental changes in a timely manner. These problems limit the further improvement and optimization of system performance. Summary of the Invention

[0004] The present invention provides an intelligent unmanned warehouse scheduling system for spare parts, comprising:

[0005] The spare parts information perception module is used to obtain comprehensive spare parts detailed information, including spare parts model, size, weight, shelf life, in and out time records, historical usage frequency, and current inventory location coordinates. The obtained spare parts detailed information is then transmitted to the demand forecasting and analysis module and the system database.

[0006] The demand forecasting and analysis module receives detailed information on spare parts transmitted by the spare parts information perception module. Using data mining and analysis techniques, it combines historical demand data, production plans, and equipment operating status factors to predict future demand for spare parts and generate a demand forecast report. It also analyzes current real-time demand, determines the urgency and priority of demand, and transmits the demand forecast report and analysis results to the core module of the intelligent scheduling algorithm.

[0007] The core module of the intelligent scheduling algorithm generates the optimal scheduling strategy based on the demand forecast report and analysis results transmitted by the demand forecast and analysis module using a multi-objective optimization algorithm, and outputs the scheduling instructions to the unmanned warehouse execution and monitoring module;

[0008] The unmanned warehouse execution and monitoring module receives the scheduling instructions output by the core module of the intelligent scheduling algorithm, controls the unmanned handling equipment and shelf shuttle unmanned warehouse equipment to complete the spare parts handling, storage and in-and-out operations; at the same time, it monitors the equipment operation status, spare parts location changes, and inventory environment temperature and humidity parameters in real time, and transmits the monitoring data feedback to the spare parts information perception module.

[0009] Furthermore, the spare parts information sensing module obtains spare parts model information in the following steps:

[0010] High-definition image recognition equipment is used to collect model identification images on the surface of spare parts. The images are processed and feature extracted through image recognition algorithms, and matched with the preset model template library to determine the spare part model and transmit the model information to the system database.

[0011] Furthermore, the spare parts information sensing module acquires spare parts size information in the following steps:

[0012] A laser ranging sensor is used to scan around spare parts to obtain distance data of various dimensions of the spare parts. The length, width and height of the spare parts are calculated through spatial geometry algorithms, and the dimension information is transmitted to the system database.

[0013] Furthermore, the steps of predicting future demand for spare parts by the demand forecasting and analysis module are as follows:

[0014] Step 201: Establish a time series prediction model. The formula of the time series prediction model is:

[0015]

[0016] Among them, F t+h represents the demand forecast value for the hth time period in the future, t represents the current time period, p and q are the autoregressive order and moving average order respectively, α i and β j is the model parameter, D t-i+1 is the historical demand data, ∈ t-j+1 is the historical forecast error;

[0017] Using the time series forecasting model, input historical demand data and related parameters to obtain preliminary demand forecast values;

[0018] Step 202: Adjust the demand forecast value in combination with the production plan. The demand forecast value after adjustment in combination with the production plan is calculated as follows:

[0019]

[0020] Among them, A t+h To consider the demand forecast value after production plan adjustment, P t+h is the planned production output for the hth time period in the future, P avg is the historical average production plan output, γ is the adjustment coefficient related to the production plan;

[0021] Step 203: The demand forecast value is modified based on the equipment operating status. The calculation formula for the modified final demand forecast value is:

[0022]

[0023] Among them, R t+h is the final demand forecast value, δ is the correction coefficient related to the equipment operation status, M t+h M is the number of equipment that is expected to require maintenance or replacement of spare parts in the next h time period. total The total number of devices.

[0024] Furthermore, the demand forecasting and analysis module determines the urgency and priority of demand, including:

[0025] Set the urgency determination formula, which is:

[0026] E=ω1×U+ω2×T+ω3×C

[0027] Where E represents the urgency of the demand, U represents the impact on production or equipment operation, and the value range is 0-10. The larger the value, the greater the impact. T represents the demand response time in hours. ω1, ω2, and ω3 are weight coefficients, and ω1+ω2+ω3=1.

[0028] According to the set urgency determination formula, the urgency value E of the demand is obtained;

[0029] Priorities are determined based on the calculated urgency value E. The higher the E value, the higher the priority.

[0030] Furthermore, the intelligent scheduling algorithm core module generates an optimal scheduling strategy using a multi-objective optimization algorithm based on the demand forecast report and analysis results transmitted by the demand forecast and analysis module, including:

[0031] Step 301: Determine the inventory cost objective function, which is:

[0032]

[0033] Among them, C inv represents inventory cost, n is the number of spare parts, c i is the unit inventory cost of the i-th spare part, s i is the inventory quantity of the i-th spare part, t i is the storage time of the i-th spare part;

[0034] Step 302: Determine the scheduling efficiency objective function, which is:

[0035]

[0036] Among them, E sch represents the scheduling efficiency index, m is the number of transport tasks, d j is the transport distance of the jth transport task, v j is the speed of the equipment performing the jth handling task, l is the number of equipment operations, τ k is the time taken for the kth device operation;

[0037] Step 303: Determine the device load balancing objective function, which is:

[0038]

[0039] Among them, B eq Indicates the device load balancing degree, R is the total number of devices, θ jr The load ratio assigned to the rth device for the jth task;

[0040] Step 304: Using the non-dominated sorting genetic algorithm NSGA-II to optimize and solve the inventory cost objective function, the scheduling efficiency objective function, and the equipment load balancing objective function, to obtain the optimal scheduling strategy.

[0041] Furthermore, the non-dominated sorting genetic algorithm NSGA-II is used to optimize the inventory cost objective function, the scheduling efficiency objective function, and the equipment load balancing objective function, and the optimal scheduling strategy is obtained, including:

[0042] Step 3041: Initialize the population and randomly generate a certain number of scheduling schemes as the initial population;

[0043] Step 3042, performing non-dominated sorting on each individual in the population, and dividing the population into different levels;

[0044] Step 3043, calculating the crowding degree of individuals in each level to maintain population diversity;

[0045] Step 3044, generating a new population through tournament selection, crossover and mutation operations;

[0046] Step 3045: perform iterative optimization, repeating steps 3042 to 3044 until a preset termination condition is met and the optimal scheduling strategy is obtained;

[0047] In each iteration, each individual is evaluated for its performance in the three objective functions of inventory cost, scheduling efficiency, and equipment load balance, and these evaluation results are used for non-dominated sorting and congestion calculation; ensuring that each individual's scheduling plan optimizes scheduling efficiency and equipment load balance while minimizing inventory costs to achieve multi-objective optimization.

[0048] Furthermore, the unmanned warehouse execution and monitoring module controls the unmanned transport equipment to perform the transport task, including:

[0049] Step 401: Use the A* algorithm to plan the optimal path for the device according to the scheduling instructions. The path search formula is:

[0050] f(n)=g(n)+h(n)

[0051] Among them, f(n) is the comprehensive cost of node n, g(n) is the actual cost from the starting point to node n, and h(n) is the estimated cost from node n to the target point;

[0052] Step 402: Control the unmanned transport equipment to move along the planned path, adjusting the equipment speed and direction in real time to ensure accurate arrival at the target cargo location.

[0053] Step 403: After arriving at the target cargo location, the control device performs the picking operation and adjusts the strength and grasping position of the picking robot arm according to the weight and size information of the spare parts.

[0054] Furthermore, the unmanned warehouse execution and monitoring module monitors inventory environment parameters including:

[0055] Deploy temperature sensors and humidity sensors at different locations in the unmanned warehouse to collect temperature T and humidity H data in real time;

[0056] Set the temperature threshold range [T min ,T max ] and humidity threshold range [H min ,H max ];

[0057] When the collected temperature T exceeds [T min ,T max ] or humidity H exceeds [H min ,H max ], the alarm mechanism is triggered and the abnormal environmental parameter information is transmitted to the system management terminal.

[0058] Furthermore, it also includes a data analysis and optimization module, which collects data generated by the spare parts information perception module, demand forecasting and analysis module, intelligent scheduling algorithm core module, and unmanned warehouse execution and monitoring module. Through data analysis and mining technology, it discovers potential problems and optimization points in system operation and feeds back to the corresponding modules to continuously optimize the performance of the scheduling system.

[0059] The above-mentioned embodiments of the present invention have at least the following beneficial effects: the intelligent unmanned warehouse scheduling system, by integrating a spare parts information perception module, a demand forecasting and analysis module, an intelligent scheduling algorithm core module, and an unmanned warehouse execution and monitoring module, can achieve all-round information acquisition and real-time monitoring of spare parts, thereby improving the accuracy and response speed of inventory management. The system uses data mining and analysis technology, combined with historical demand data, production plans, and equipment operating status, to predict future demand for spare parts, generate more accurate demand forecast reports, and then formulate the optimal scheduling strategy. This intelligent scheduling method can improve the efficiency of spare parts in and out of the warehouse, reduce inventory costs, and at the same time ensure the efficiency of equipment operation and the stability of environmental parameters.

[0060] Furthermore, the system utilizes multi-objective optimization algorithms, such as the non-dominated sorting genetic algorithm (NSGA-II), which simultaneously considers multiple objectives, including inventory costs, scheduling efficiency, and equipment load balancing, enabling intelligent scheduling based on multi-objective optimization. This optimization strategy not only minimizes inventory costs but also improves scheduling efficiency and equipment utilization, thereby achieving rational allocation and efficient utilization of resources. The system's real-time monitoring capabilities ensure that equipment operating status and environmental parameters are always optimal, further enhancing system reliability and stability. Through continuous feedback and optimization from the data analysis and optimization modules, the system continuously improves itself to adapt to changing production needs and environmental conditions, thereby achieving continuous performance improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily apparent by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present invention are shown by way of example and not limitation, in which:

[0062] Figure 1 A schematic structural diagram of an intelligent unmanned warehouse scheduling system for spare parts provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0063] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided solely to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. Rather, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0064] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a system, apparatus, device, method, or computer program product. Therefore, the present invention may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of hardware and software.

[0065] It should be noted that any number of elements in the drawings is for illustration only and not for limitation, and any naming is only for distinction and does not have any limiting meaning.

[0066] Reference below Figure 1 , Figure 1 This is a schematic diagram of the structure of an intelligent unmanned warehouse scheduling system for spare parts provided by one embodiment of the present invention. Figure 1 As shown, an intelligent unmanned warehouse scheduling system 100 for spare parts includes:

[0067] The spare parts information sensing module 101 is used to obtain detailed information about spare parts in all aspects, including spare parts model, size, weight, shelf life, entry and exit time records, historical usage frequency, and current inventory location coordinates, and transmit the obtained spare parts detailed information to the demand forecasting and analysis module and the system database;

[0068] The demand forecasting and analysis module 102 receives detailed information on spare parts transmitted by the spare parts information sensing module. Using data mining and analysis techniques, it combines historical demand data, production plans, and equipment operating status factors to forecast future demand for spare parts and generate a demand forecast report. It also analyzes current real-time demand, determines the urgency and priority of demand, and transmits the demand forecast report and analysis results to the intelligent scheduling algorithm core module.

[0069] The intelligent scheduling algorithm core module 103 generates an optimal scheduling strategy using a multi-objective optimization algorithm based on the demand forecast report and analysis results transmitted by the demand forecast and analysis module, and outputs the scheduling instructions to the unmanned warehouse execution and monitoring module;

[0070] The unmanned warehouse execution and monitoring module 104 receives the scheduling instructions output by the intelligent scheduling algorithm core module, controls the unmanned handling equipment and shelf shuttle unmanned warehouse equipment to complete the spare parts handling, storage and in-and-out operations; at the same time, it monitors the equipment operation status, spare parts location changes, and inventory environment temperature and humidity parameters in real time, and transmits the monitoring data feedback to the spare parts information perception module.

[0071] It's important to note that the system includes a spare parts information perception module, which is responsible for obtaining comprehensive spare parts information. This information includes spare parts model, size, weight, expiration date, entry and exit times, historical usage frequency, and current inventory location coordinates. The spare parts information perception module is the system's data collection core, ensuring the system accurately understands the real-time status of spare parts in inventory. Spare parts, in this context, refer to the various accessories and materials required to replace or repair equipment.

[0072] Specifically, the spare part information perception module uses high-definition image recognition equipment and laser ranging sensors to obtain spare part model and size information. The high-definition image recognition equipment captures images of the model identification on the spare part's surface, processes them using an image recognition algorithm, and matches them against a library of preset model templates to determine the spare part's model. The laser ranging sensor then scans the spare part, acquiring distance data in various dimensions and calculating the spare part's length, width, and height using a spatial geometry algorithm. These parameters ensure the system can accurately identify and measure key attributes of spare parts.

[0073] Preferably, the spare parts information sensing module can also include a weight sensor and an RFID tag reader to obtain the weight and electronic tag information of the spare parts. The weight sensor can be installed on the shelf or conveyor belt to monitor the weight of the spare parts in real time; the RFID tag reader can quickly read the electronic tag on the spare part to obtain its unique identifier and related attributes.

[0074] Furthermore, the system can also integrate environmental monitoring sensors, such as temperature and humidity sensors, to monitor the inventory environment and ensure that spare parts are stored in the required conditions. These additional monitoring methods can further improve the accuracy and reliability of the system.

[0075] In some embodiments, the spare part information sensing module acquires spare part model information by:

[0076] High-definition image recognition equipment is used to collect model identification images on the surface of spare parts. The images are processed and feature extracted through image recognition algorithms, and matched with the preset model template library to determine the spare part model and transmit the model information to the system database.

[0077] It should be noted that the system obtains spare part model information through the spare part information perception module. Specifically, this involves using high-definition image recognition equipment to capture images of the model identification on the spare part's surface. This image is then processed and feature extracted using an image recognition algorithm. This is then matched against a preset model template library to determine the spare part model and transmit this information to the system database. High-definition image recognition equipment, which captures high-resolution images, is crucial for accurately identifying spare part models.

[0078] Specifically, images captured by high-definition image recognition equipment are processed using an image recognition algorithm that can identify and extract key features within the image, such as text, symbols, and patterns. A preset model template library contains a series of standard images of known spare part models, which are used for comparison and matching with the captured images. The matching process involves identifying and comparing feature points to determine the specific model of the spare part. The system database is used to store this model information for subsequent demand forecasting and scheduling.

[0079] Preferably, the image recognition algorithm can utilize deep learning technology to improve recognition accuracy and robustness through model training. Furthermore, the system can be equipped with multiple high-definition image recognition devices to capture images of spare parts from different angles to enhance comprehensiveness and accuracy. The library of preset model templates can be regularly updated and expanded to include more spare part models.

[0080] Furthermore, as an alternative, the system can also use a barcode or QR code scanner to obtain spare part model information. This solution can serve as an effective supplement, especially when image recognition conditions are not ideal.

[0081] In some embodiments, the spare parts information sensing module acquires spare parts size information by:

[0082] A laser ranging sensor is used to scan around spare parts to obtain distance data of various dimensions of the spare parts. The length, width and height of the spare parts are calculated through spatial geometry algorithms, and the dimension information is transmitted to the system database.

[0083] It should be noted that the system obtains spare part dimensional information through the spare part information perception module. Specifically, a laser ranging sensor scans the spare part, acquiring distance data in each dimension. Using spatial geometry algorithms, the system calculates the spare part's length, width, and height, and transmits this dimensional information to the system database. A laser ranging sensor, in this context, uses laser technology to measure distances, accurately capturing the three-dimensional dimensions of spare parts.

[0084] Specifically, laser ranging sensors measure distance by emitting laser pulses and receiving reflected light. These sensors can be mounted on robots or automated equipment to perform a 360-degree scan around spare parts, acquiring comprehensive dimensional data. Spatial geometry algorithms use this distance data to calculate the spare part's three-dimensional dimensions using mathematical models. The system database stores this dimensional information for subsequent demand forecasting and scheduling.

[0085] Ideally, the laser rangefinder can be equipped with a high-resolution scanning head to improve measurement accuracy and speed. Furthermore, the system can employ multiple laser rangefinders to scan from different angles to ensure more comprehensive and accurate dimensional data. Regarding the choice of spatial geometry algorithms, more advanced 3D reconstruction algorithms, such as point cloud processing technology, can be employed to enhance the accuracy of dimensional calculations.

[0086] Furthermore, as an alternative, the system can also consider using structured light scanning technology or a stereo vision system to obtain the size information of spare parts. These technologies can provide additional flexibility and robustness in different application scenarios.

[0087] In some embodiments, the demand forecasting and analysis module predicts future demand for spare parts in the following steps:

[0088] Step 201: Establish a time series prediction model. The formula of the time series prediction model is:

[0089]

[0090] Among them, F t+h represents the demand forecast value for the hth time period in the future, t represents the current time period, p and q are the autoregressive order and moving average order respectively, α i and β j is the model parameter, D t-i+1 is the historical demand data, ∈ t-j+1 is the historical forecast error;

[0091] Using the time series forecasting model, input historical demand data and related parameters to obtain preliminary demand forecast values;

[0092] Step 202: Adjust the demand forecast value in combination with the production plan. The demand forecast value after adjustment in combination with the production plan is calculated as follows:

[0093]

[0094] Among them, A t+h To consider the demand forecast value after production plan adjustment, P t+h is the planned production output for the hth time period in the future, Pavg is the historical average production plan output, γ is the adjustment coefficient related to the production plan;

[0095] Step 203: The demand forecast value is modified based on the equipment operating status. The calculation formula for the modified final demand forecast value is:

[0096]

[0097] Among them, R t+h is the final demand forecast value, δ is the correction coefficient related to the equipment operation status, M t+h M is the number of equipment that is expected to require maintenance or replacement of spare parts in the next h time period. total The total number of devices.

[0098] It's important to note that the system's demand forecasting and analysis module uses a time series forecasting model to predict future demand for spare parts. A time series forecasting model is a statistical method used to predict future values based on historical data. In this process, model parameters, autoregressive order, and moving average order are key factors that determine the model's predictive power. Here, autoregressive order and moving average order refer to parameters used in time series analysis to describe the autocorrelation of data.

[0099] Specifically, the formula for a time series forecasting model involves autoregressive and moving average terms, which use historical demand data and historical forecast errors to calculate future demand forecasts. Model parameters, including autoregressive and moving average coefficients, are estimated from historical data using statistical analysis methods such as maximum likelihood estimation or least squares. Historical demand data represents actual demand for spare parts over a past period of time, while historical forecast errors represent the difference between past forecasts and actual values.

[0100] Preferably, the time series forecasting model can adopt the ARIMA (Autoregressive Integrated Moving Average) model, which is a widely used time series forecasting method that is particularly suitable for processing non-stationary time series data. In practical applications, the selection of the autoregressive order and the moving average order can be optimized using information criteria such as the Akaike Information Criterion (AIC) or the Bayesian Information Criterion (BIC).

[0101] Furthermore, the model can also include seasonal factors to more accurately capture seasonal variations in demand. As an alternative, you can consider using machine learning methods such as random forests or neural networks to build demand forecast models. These methods are able to handle more complex nonlinear relationships and may provide higher forecast accuracy.

[0102] In some embodiments, the demand forecasting and analysis module determines the urgency and priority of demand including:

[0103] Set the urgency determination formula, which is:

[0104] E=ω1×U+ω2×T+ω3×C

[0105] Where E represents the urgency of the demand, U represents the impact on production or equipment operation, and the value range is 0-10. The larger the value, the greater the impact. T represents the demand response time in hours. ω1, ω2, and ω3 are weight coefficients, and ω1+ω2+ω3=1.

[0106] According to the set urgency determination formula, the urgency value E of the demand is obtained;

[0107] Priorities are determined based on the calculated urgency value E. The higher the E value, the higher the priority.

[0108] It's important to note that the system's demand forecasting and analysis module also includes steps for determining demand urgency and priority. The urgency determination formula is a mathematical model used to quantify demand urgency, taking into account the impact on production or equipment operation, the demand response time, and the corresponding weighting coefficient. Here, the demand urgency value is a numerical value that represents the urgency of the demand, while the weighting coefficient is a parameter used to adjust the importance of different factors in the urgency determination.

[0109] Specifically, the demand response time in the urgency determination formula refers to the time interval from the identification of the demand to the start of the response action, usually in hours. The impact degree is a score used to measure the impact that the failure to meet the demand in a timely manner may have on production or equipment operation. The value range is 0 to 10, and the larger the value, the greater the impact. Weight coefficient 、 、 It is an indicator of the relative importance assigned to the impact, demand response time and other relevant factors (such as inventory levels). Their sum is 1, indicating the comprehensive consideration of these factors in the urgency determination.

[0110] Preferably, the urgency determination formula can be adjusted according to actual business needs and historical data analysis results. For example, if it is found that the demand response time has a greater impact on production, the urgency determination formula can be appropriately increased. value.

[0111] Furthermore, the system can also introduce more factors, such as inventory levels and supply chain stability, to further improve the urgency determination formula. As an alternative, machine learning methods such as logistic regression or decision trees can be used to automatically determine the urgency and priority of demand. These methods can learn from large amounts of historical data and predict the urgency of demand.

[0112] In some embodiments, the intelligent scheduling algorithm core module generates an optimal scheduling strategy using a multi-objective optimization algorithm based on the demand forecast report and analysis results transmitted by the demand forecast and analysis module, including:

[0113] Step 301: Determine the inventory cost objective function, which is:

[0114]

[0115] Among them, C inv represents inventory cost, n is the number of spare parts, c i is the unit inventory cost of the i-th spare part, s i is the inventory quantity of the i-th spare part, t i is the storage time of the i-th spare part;

[0116] Step 302: Determine the scheduling efficiency objective function, which is:

[0117]

[0118] Among them, E sch represents the scheduling efficiency index, m is the number of transport tasks, d j is the transport distance of the jth transport task, v j is the speed of the equipment performing the jth handling task, l is the number of equipment operations, τ k is the time taken for the kth device operation;

[0119] Step 303: Determine the device load balancing objective function, which is:

[0120]

[0121] Among them, B eq Indicates the device load balancing degree, R is the total number of devices, θ jr The load ratio assigned to the rth device for the jth task;

[0122] Step 304: Using the non-dominated sorting genetic algorithm NSGA-II to optimize and solve the inventory cost objective function, the scheduling efficiency objective function, and the equipment load balancing objective function, to obtain the optimal scheduling strategy.

[0123] It should be noted that the core module of the intelligent scheduling algorithm uses a multi-objective optimization algorithm to generate the optimal scheduling strategy based on demand forecast reports and analysis results. A multi-objective optimization algorithm is a mathematical method used to find the best solution among multiple conflicting objectives. In this context, a multi-objective optimization algorithm refers to an algorithm that can simultaneously consider multiple optimization objectives, such as inventory cost, scheduling efficiency, and equipment load balancing.

[0124] Specifically, the inventory cost objective function considers the unit inventory cost, inventory quantity, and inventory time of spare parts; the scheduling efficiency objective function considers the number of handling tasks, handling distance, equipment speed, and number of equipment operations; and the equipment load balancing objective function focuses on the load ratio of tasks allocated to different equipment. The parameters of these objective functions need to be determined based on actual business needs and historical data analysis to ensure that the scheduling strategy is both economical and efficient.

[0125] The non-dominated sorting genetic algorithm (NSGA-II) is preferably used as a specific implementation of a multi-objective optimization algorithm. When initializing the population, a certain number of scheduling solutions can be randomly generated as the initial population. During the non-dominated sorting process, the population can be divided into different levels, and the crowding degree of individuals in each level is calculated to maintain population diversity. New populations are generated through tournament selection, crossover, and mutation operations, and iterative optimization is performed until the preset termination condition is met.

[0126] More specifically, the individual evaluation process can be refined to ensure that each individual scheduling plan optimizes scheduling efficiency and equipment load balancing while minimizing inventory costs, so as to achieve multi-objective optimization.

[0127] Furthermore, the parameters of the non-dominated sorting genetic algorithm (NSGA-II), such as the crossover rate and mutation rate, can be adjusted to adapt to different scheduling scenarios. Furthermore, other multi-objective optimization algorithms, such as multi-objective particle swarm optimization (MOPSO) or multi-objective simulated annealing (MOSA), can be considered to provide more algorithm choices and comparisons, thereby finding the scheduling strategy that best suits current business needs.

[0128] In some embodiments, the non-dominated sorting genetic algorithm NSGA-II is used to optimize and solve the inventory cost objective function, the scheduling efficiency objective function, and the equipment load balancing objective function to obtain the optimal scheduling strategy including:

[0129] Step 3041: Initialize the population and randomly generate a certain number of scheduling schemes as the initial population;

[0130] Step 3042, performing non-dominated sorting on each individual in the population, and dividing the population into different levels;

[0131] Step 3043, calculating the crowding degree of individuals in each level to maintain population diversity;

[0132] Step 3044, generating a new population through tournament selection, crossover and mutation operations;

[0133] Step 3045: perform iterative optimization, repeating steps 3042 to 3044 until a preset termination condition is met and the optimal scheduling strategy is obtained;

[0134] In each iteration, each individual is evaluated for its performance in the three objective functions of inventory cost, scheduling efficiency, and equipment load balance, and these evaluation results are used for non-dominated sorting and congestion calculation; ensuring that each individual's scheduling plan optimizes scheduling efficiency and equipment load balance while minimizing inventory costs to achieve multi-objective optimization.

[0135] It should be noted that the system uses the non-dominated sorting genetic algorithm (NSGA-II) to optimize the inventory cost objective function, the scheduling efficiency objective function, and the equipment load balancing objective function to obtain the optimal scheduling strategy. The non-dominated sorting genetic algorithm (NSGA-II) is an evolutionary algorithm used to solve multi-objective optimization problems. It iteratively improves the solution by simulating natural selection and genetic mechanisms. In this context, non-dominated sorting refers to a sorting method that stratifies the population based on the dominance relationship of individuals in the objective function.

[0136] Specifically, the implementation of the non-dominated sorting genetic algorithm (NSGA-II) includes the following steps: initializing the population, performing non-dominated sorting, calculating the congestion degree, and performing selection, crossover, and mutation operations. Initializing the population involves randomly generating a certain number of scheduling solutions as initial solutions. Non-dominated sorting divides individuals in the population into different levels based on their non-dominated relationships in the objective function. Crowding degree calculation is used to assess the distance between individuals in the population to maintain population diversity. Selection, crossover, and mutation are the basic genetic operations in genetic algorithms and are used to generate new populations.

[0137] Preferably, the implementation of the non-dominated sorting genetic algorithm (NSGA-II) can be further refined. For example, when initializing the population, a specific encoding strategy can be used to represent the scheduling scheme, such as binary encoding or real number encoding.

[0138] Furthermore, in the selection operation, tournament selection methods can be used, in which a certain number of individuals are randomly selected for comparison, and the best individuals are selected for the next generation. In the crossover operation, methods such as simulated two-point crossover or uniform crossover can be used to combine two parent individuals to generate a new offspring. In the mutation operation, techniques such as random perturbation or crossover mutation can be used to introduce new genetic variation and increase population diversity.

[0139] Furthermore, the algorithm parameters, such as crossover rate, mutation rate, and population size, can be set to adapt to different optimization problems and solution environments. As an alternative, other multi-objective optimization algorithms, such as multi-objective particle swarm optimization (MOPSO) or multi-objective ant colony optimization (MOACO), can be considered. These algorithms provide different search strategies and optimization mechanisms and may provide better performance in specific situations.

[0140] In some embodiments, the unmanned warehouse execution and monitoring module controls the unmanned transport equipment to perform the transport task, including:

[0141] Step 401: Use the A* algorithm to plan the optimal path for the device according to the scheduling instructions. The path search formula is:

[0142] f(n)=g(n)+h(n)

[0143] Among them, f(n) is the comprehensive cost of node n, g(n) is the actual cost from the starting point to node n, and h(n) is the estimated cost from node n to the target point;

[0144] Step 402: Control the unmanned transport equipment to move along the planned path, adjusting the equipment speed and direction in real time to ensure accurate arrival at the target cargo location.

[0145] Step 403: After arriving at the target cargo location, the control device performs the picking operation and adjusts the strength and grasping position of the picking robot arm according to the weight and size information of the spare parts.

[0146] It's important to note that the unmanned warehouse execution and monitoring module controls the unmanned transport equipment to perform its transport tasks, including using the A* algorithm to plan the equipment's optimal path. The A* algorithm is a heuristic search algorithm used to find the shortest path from a starting point to a destination in a graph. Here, the A* algorithm refers to a path search method that combines the advantages of best-first search and Dijkstra's algorithm. It determines the optimal path by evaluating actual and estimated costs.

[0147] Specifically, the path search formula in the A* algorithm is f(n) = g(n) + h(n), where g(n) is the actual cost from the starting point to node n, and h(n) is the estimated cost from node n to the destination. The actual cost is calculated based on the paths previously traveled by the unmanned transport device, while the estimated cost is typically predicted using a preset heuristic function, such as Manhattan distance or Euclidean distance. Parameter settings, including the choice of heuristic function and the assignment of weights, directly affect the efficiency and accuracy of the path search.

[0148] The implementation of the A* algorithm can be further refined. For example, when planning a path, the heuristic function h(n) can be adjusted to more accurately reflect obstacles and terrain characteristics in the actual environment. In addition, a dynamic path adjustment mechanism can be introduced to quickly replan the path when the environment changes or when encountering unknown obstacles.

[0149] Furthermore, when controlling unmanned transport equipment, more sophisticated speed and direction control strategies, such as PID control or fuzzy logic control, can be adopted to ensure that the equipment can reach the target cargo location smoothly and accurately.

[0150] Furthermore, as an alternative, other path planning algorithms, such as the Dijkstra algorithm or the RRT (Rapidly Exploring Random Trees) algorithm, may be considered. These algorithms may provide different performance characteristics in different application scenarios and environmental conditions.

[0151] In some embodiments, the unmanned warehouse execution and monitoring module monitors inventory environment parameters including:

[0152] Deploy temperature sensors and humidity sensors at different locations in the unmanned warehouse to collect temperature T and humidity H data in real time;

[0153] Set the temperature threshold range [T min ,T max ] and humidity threshold range [H min ,H max ];

[0154] When the collected temperature T exceeds [T min ,T max ] or humidity H exceeds [H min ,H max ], the alarm mechanism is triggered and the abnormal environmental parameter information is transmitted to the system management terminal.

[0155] It's important to note that the unmanned warehouse execution and monitoring module is responsible for monitoring inventory environmental parameters. This involves deploying temperature and humidity sensors at various locations within the warehouse to collect real-time temperature and humidity data. These sensors are devices that can detect changes in ambient temperature and humidity, and are crucial for maintaining inventory stability and ensuring the proper storage conditions for spare parts.

[0156] Specifically, temperature and humidity sensors can be deployed at key locations in unmanned warehouses, such as warehouse entrances, exits, and cargo storage areas, to ensure comprehensive monitoring of environmental conditions. These sensors can be configured with different threshold ranges, such as temperature and humidity, to define the normal fluctuation range of environmental parameters.

[0157] Furthermore, parameter setting involves determining the specific values of these thresholds, which are typically set based on the storage requirements and safety standards of the spare parts. For example, certain sensitive electronic spare parts may need to be stored within specific temperature and humidity ranges to prevent damage.

[0158] Preferably, the monitoring system can further refine its operational procedures, including setting specific alarm mechanisms. When collected temperature or humidity data exceeds a preset threshold, the system triggers an alarm and transmits information about the environmental parameter anomaly to the system management terminal. This alarm mechanism can take the form of an audible or visual alarm, SMS notification, or system log recording.

[0159] More specifically, the system can employ intelligent data analysis techniques, such as trend analysis and anomaly detection algorithms, to predict potential environmental risks and provide early warnings before problems occur. As an alternative, wireless sensor networks can be considered, as they offer more flexible deployment and higher data collection frequency, especially in large warehouses or multi-area monitoring scenarios.

[0160] In some embodiments, a data analysis and optimization module is also included to collect data generated by the spare parts information perception module, the demand forecasting and analysis module, the intelligent scheduling algorithm core module, and the unmanned warehouse execution and monitoring module. Through data analysis and mining technology, potential problems and optimization points in the system operation are discovered and fed back to the corresponding modules to continuously optimize the performance of the scheduling system.

[0161] It's important to note that the system also includes a data analysis and optimization module, which collects data generated by each module and uses data analysis and mining techniques to identify potential issues and optimization points in system operation. Here, the data analysis and optimization module refers to an integrated system component responsible for conducting in-depth analysis of collected data to identify opportunities for improving system performance.

[0162] Specifically, the data analysis and optimization module collects data from the spare parts information perception module, demand forecasting and analysis module, intelligent scheduling algorithm core module, and unmanned warehouse execution and monitoring module. This data includes detailed information on spare parts, demand forecast reports, scheduling strategies, and execution monitoring results.

[0163] Furthermore, data analysis and mining techniques can include statistical analysis, machine learning algorithms, pattern recognition, and other methods to extract valuable information from large amounts of data. Parameter setting may involve selecting appropriate algorithms, setting algorithm parameters, and determining data preprocessing steps, all of which need to be customized based on specific business needs and data characteristics.

[0164] The operational steps of the data analysis and optimization module can be further refined. For example, regular data quality checks can be implemented to ensure the accuracy of collected data. Furthermore, customized reports and dashboards can be developed to intuitively display analysis results and optimization recommendations. In implementing optimization strategies, automated feedback mechanisms can be used to directly apply analysis results to system configuration adjustments, achieving self-optimization.

[0165] More specifically, predictive maintenance algorithms can be introduced to predict potential failures and maintenance needs based on equipment operating data, thereby reducing downtime and improving system reliability. As an alternative, cloud service platforms can be considered for storing and analyzing data, which can provide higher computing power and flexibility, especially when processing large-scale data sets.

[0166] The above-mentioned embodiments of the present invention have the following beneficial effects: The intelligent unmanned warehouse scheduling system described in the present invention can achieve all-round information acquisition and real-time monitoring of spare parts through comprehensive spare parts information perception, demand forecast analysis, intelligent scheduling algorithm core, and unmanned warehouse execution and monitoring module, thereby improving the accuracy and response speed of inventory management. The system uses data mining and analysis technology, combined with historical demand data, production plans and equipment operating status, to predict future demand for spare parts, generate more accurate demand forecast reports, and then formulate the optimal scheduling strategy. This intelligent scheduling method can improve the efficiency of spare parts in and out of the warehouse, reduce inventory costs, and at the same time ensure the efficiency of equipment operation and the stability of environmental parameters.

[0167] Furthermore, the system utilizes multi-objective optimization algorithms, such as the non-dominated sorting genetic algorithm (NSGA-II), which simultaneously considers multiple objectives, including inventory costs, scheduling efficiency, and equipment load balancing, enabling intelligent scheduling based on multi-objective optimization. This optimization strategy not only minimizes inventory costs but also improves scheduling efficiency and equipment utilization, achieving rational allocation and efficient utilization of resources. The system's real-time monitoring capabilities ensure that equipment operating status and environmental parameters are always optimal, further enhancing system reliability and stability. Through continuous feedback and optimization from the data analysis and optimization modules, the system continuously improves itself to adapt to changing production needs and environmental conditions, achieving continuous performance improvement.

[0168] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or a terminal device such as a computer, a server, a mobile phone, or a tablet.

[0169] The above descriptions are merely some preferred embodiments of the present invention and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also encompass other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned inventive concept. For example, a technical solution formed by mutually replacing the above-mentioned features with (but not limited to) technical features having similar functions disclosed in the embodiments of the present invention.

Claims

1. An intelligent unmanned warehouse scheduling system for spare parts, characterized in that: include: The spare parts information perception module is used to obtain comprehensive spare parts detailed information, including spare parts model, size, weight, shelf life, in and out time records, historical usage frequency, and current inventory location coordinates. The obtained spare parts detailed information is then transmitted to the demand forecasting and analysis module and the system database. The demand forecasting and analysis module receives detailed information on spare parts transmitted by the spare parts information perception module. Using data mining and analysis techniques, it combines historical demand data, production plans, and equipment operating status factors to predict future demand for spare parts and generate a demand forecast report. It also analyzes current real-time demand, determines the urgency and priority of demand, and transmits the demand forecast report and analysis results to the core module of the intelligent scheduling algorithm. The core module of the intelligent scheduling algorithm generates the optimal scheduling strategy based on the demand forecast report and analysis results transmitted by the demand forecast and analysis module using a multi-objective optimization algorithm, and outputs the scheduling instructions to the unmanned warehouse execution and monitoring module; The unmanned warehouse execution and monitoring module receives the scheduling instructions output by the core module of the intelligent scheduling algorithm, controls the unmanned handling equipment and shelf shuttle unmanned warehouse equipment to complete the spare parts handling, storage and in-and-out operations; at the same time, it monitors the equipment operation status, spare parts location changes, and inventory environment temperature and humidity parameters in real time, and transmits the monitoring data feedback to the spare parts information perception module.

2. The intelligent unmanned warehouse scheduling system for spare parts according to claim 1 is characterized in that: The steps for the spare parts information sensing module to obtain spare parts model information are as follows: High-definition image recognition equipment is used to collect model identification images on the surface of spare parts. The images are processed and feature extracted through image recognition algorithms, and matched with the preset model template library to determine the spare part model and transmit the model information to the system database.

3. The intelligent unmanned warehouse scheduling system for spare parts according to claim 1 is characterized in that: The steps for the spare parts information sensing module to obtain spare parts size information are as follows: A laser ranging sensor is used to scan around spare parts to obtain distance data of various dimensions of the spare parts. The length, width and height of the spare parts are calculated through spatial geometry algorithms, and the dimension information is transmitted to the system database.

4. The intelligent unmanned warehouse scheduling system for spare parts according to claim 1 is characterized in that: The steps of the demand forecasting and analysis module for predicting future demand for spare parts are as follows: Step 201: Establish a time series prediction model. The formula of the time series prediction model is: Among them, F t+h represents the demand forecast value for the hth time period in the future, t represents the current time period, p and q are the autoregressive order and moving average order respectively, α i and β j is the model parameter, D t-i+1 is the historical demand data, ∈ t-j+1 is the historical forecast error; Using the time series forecasting model, input historical demand data and related parameters to obtain preliminary demand forecast values; Step 202: Adjust the demand forecast value in combination with the production plan. The demand forecast value after adjustment in combination with the production plan is calculated as follows: Among them, A t+h To consider the demand forecast value after production plan adjustment, P t+h is the planned production output for the hth time period in the future, P avg is the historical average production plan output, γ is the adjustment coefficient related to the production plan; Step 203: The demand forecast value is modified based on the equipment operating status. The calculation formula for the modified final demand forecast value is: Among them, R t+h is the final demand forecast value, δ is the correction coefficient related to the equipment operation status, M t+h M is the number of equipment that is expected to require maintenance or replacement of spare parts in the next h time period. total is the total number of devices.

5. The intelligent unmanned warehouse scheduling system for spare parts according to claim 1 is characterized in that: The demand forecasting and analysis module determines the urgency and priority of demand by: Set the urgency determination formula, which is: E=ω1×U+ω2×T+ω3×C Where E represents the urgency of the demand, U represents the impact on production or equipment operation, and the value range is 0-10. The larger the value, the greater the impact. T represents the demand response time in hours. ω1, ω2, and ω3 are weight coefficients, and ω1+ω2+ω3=1. According to the set urgency determination formula, the urgency value E of the demand is obtained; Priorities are determined based on the calculated urgency value E. The higher the E value, the higher the priority.

6. The intelligent unmanned warehouse scheduling system for spare parts according to claim 1 is characterized in that: The core module of the intelligent scheduling algorithm uses a multi-objective optimization algorithm to generate the optimal scheduling strategy based on the demand forecast report and analysis results transmitted by the demand forecast and analysis module, including: Step 301: Determine the inventory cost objective function, which is: Among them, C inv represents inventory cost, n is the number of spare parts, c i is the unit inventory cost of the i-th spare part, s i is the inventory quantity of the i-th spare part, t i is the storage time of the i-th spare part; Step 302: Determine the scheduling efficiency objective function, which is: Among them, E sch represents the scheduling efficiency index, m is the number of transport tasks, d j is the transport distance of the jth transport task, v j is the speed of the equipment performing the jth handling task, l is the number of equipment operations, τ k is the time taken for the kth device operation; Step 303: Determine the device load balancing objective function, which is: Among them, B eq Indicates the device load balancing degree, R is the total number of devices, θ jr The load ratio assigned to the rth device for the jth task; Step 304: Using the non-dominated sorting genetic algorithm NSGA-II to optimize and solve the inventory cost objective function, the scheduling efficiency objective function, and the equipment load balancing objective function, to obtain the optimal scheduling strategy.

7. The system according to claim 6, characterized in that The non-dominated sorting genetic algorithm NSGA-II is used to optimize the inventory cost objective function, the scheduling efficiency objective function, and the equipment load balancing objective function, and the optimal scheduling strategy is obtained, including: Step 3041: Initialize the population and randomly generate a certain number of scheduling schemes as the initial population; Step 3042, performing non-dominated sorting on each individual in the population, and dividing the population into different levels; Step 3043, calculating the crowding degree of individuals in each level to maintain population diversity; Step 3044, generating a new population through tournament selection, crossover and mutation operations; Step 3045: perform iterative optimization, repeating steps 3042 to 3044 until a preset termination condition is met and the optimal scheduling strategy is obtained; In each iteration, each individual is evaluated for its performance in the three objective functions of inventory cost, scheduling efficiency, and equipment load balance, and these evaluation results are used for non-dominated sorting and congestion calculation; ensuring that each individual's scheduling plan optimizes scheduling efficiency and equipment load balance while minimizing inventory costs to achieve multi-objective optimization.

8. The intelligent unmanned warehouse scheduling system for spare parts according to claim 1 is characterized in that: The unmanned warehouse execution and monitoring module controls the unmanned transport equipment to perform the transport task, including: Step 401: Use the A* algorithm to plan the optimal path for the device according to the scheduling instructions. The path search formula is: f(n)=g(n)+h(n) Among them, f(n) is the comprehensive cost of node n, g(n) is the actual cost from the starting point to node n, and h(n) is the estimated cost from node n to the target point; Step 402: Control the unmanned transport equipment to move along the planned path, adjusting the equipment speed and direction in real time to ensure accurate arrival at the target cargo location. Step 403: After arriving at the target cargo location, the control device performs the picking operation and adjusts the strength and grasping position of the picking robot arm according to the weight and size information of the spare parts.

9. The intelligent unmanned warehouse scheduling system for spare parts according to claim 1, characterized in that: The unmanned warehouse execution and monitoring module monitors inventory environment parameters including: Deploy temperature sensors and humidity sensors at different locations in the unmanned warehouse to collect temperature T and humidity H data in real time; Set the temperature threshold range [T min ,T max ] and humidity threshold range [H min ,H max ]; When the collected temperature T exceeds [T min ,T max ] or humidity H exceeds [H min ,H max ], the alarm mechanism is triggered and the abnormal environmental parameter information is transmitted to the system management terminal.

10. The intelligent unmanned warehouse scheduling system for spare parts according to claim 1, characterized in that: It also includes a data analysis and optimization module, which collects data generated by the spare parts information perception module, demand forecasting and analysis module, intelligent scheduling algorithm core module, and unmanned warehouse execution and monitoring module. Through data analysis and mining technology, it discovers potential problems and optimization points in system operation and feeds back to the corresponding modules to continuously optimize the performance of the scheduling system.

Images