Ship rope production scheduling method and scheduling system

Abstract

The invention discloses a ship rope production scheduling method and scheduling system, and relates to the technical field of ship rope production, a production progress monitoring module can accurately master the production rate change per hour and generate an instant production progress index Pi by monitoring the equipment state and the production batch progress in real time. The data supports the real-time adjustment of a production plan, and ensures that the resource allocation and time arrangement are more reasonable in the production process, thereby improving the production efficiency and reducing the delay. Meanwhile, the system can predict the future production trend based on the change of the production schedule index, thereby facilitating the adjustment of the production strategy and avoiding the phenomenon of production bottleneck or shutdown. And the consumption analysis and feature extraction module can extract consumption parameters according to the current production progress and the production progress index, and accurately monitor the material consumption. By analyzing and extracting consumption characteristics, the system can identify whether an abnormal material consumption condition exists in the production process, so that potential production problems can be found in advance.

Description

Technical Field

[0001] The invention relates to the technical field of ship rope production, and in particular to a ship rope production scheduling method and a scheduling system. Background Art

[0002] In the field of shipbuilding and marine engineering, ropes and cables are a key basic product, widely used in ship mooring, anchoring, towing and other scenarios. The production process of ropes and cables involves multiple links such as material procurement, production scheduling, quality inspection and logistics management, and the core function of the production scheduling system is to ensure the efficient and coordinated operation of these links. With the diversification and personalization of demand, the traditional manual scheduling method is difficult to cope with complex production scenarios, and it is urgent to introduce an intelligent scheduling system.

[0003] At present, the material demand forecasting and production scheduling of ship rope production mainly rely on experience or static rules, which are difficult to cope with the challenges of market demand fluctuations and personalized customization. This static management method is prone to a series of problems when facing complex demands: material demand cannot accurately match the actual production plan, excessive purchases may lead to inventory backlogs, or production interruptions may occur due to material shortages. In addition, the dispersion of production data and the lack of intelligent analysis methods make it difficult for the scheduling system to optimize resource allocation in real time, and cannot meet the needs of efficient, accurate, and dynamic scheduling. As a result, production efficiency, resource utilization, and corporate competitiveness are all restricted.

[0004] The above problems mainly arise from the fact that the traditional scheduling model lacks the ability to comprehensively analyze data and respond to real-time dynamic changes. On the one hand, historical order data, production progress information, and raw material consumption are often stored independently, and fail to be deeply integrated and intelligently analyzed on a unified platform; on the other hand, the lack of efficient prediction algorithms makes it impossible to dynamically identify potential trends in production demand and material usage. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a ship cable production scheduling method and scheduling system, which solve the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a ship cable production scheduling system, including a historical data acquisition module, a production progress monitoring module, a consumption analysis and feature extraction module, a material demand forecasting module, a procurement scheduling and inventory management module and a feedback and optimization module;

[0007] The historical data collection module collects past order data, material usage records, production cycle data, production batch fluctuation coefficient Bp and material cycle peak intensity Mp, and performs cleaning and standardization processing to integrate them into a data set W;

[0008] The production progress monitoring module monitors the current production progress in real time according to the data set W, including the equipment operation status and the production batch completion progress, and records the hourly production rate changes to obtain the real-time production progress index Pi;

[0009] The consumption analysis and feature extraction module analyzes the current production progress and the production progress index Pi, and extracts consumption parameters, including the average consumption rate mc and the consumption abnormality coefficient ae;

[0010] The material demand forecasting module predicts future material demand through a long short-term memory network model based on the acquired consumption parameters to obtain future material demand Dfu;

[0011] The purchasing scheduling and inventory management module automatically adjusts the material purchasing plan and inventory strategy according to the future material demand Dfu;

[0012] After the material procurement plan and inventory strategy are executed, the feedback and optimization module collects production and procurement data, evaluates the prediction accuracy, obtains the prediction deviation coefficient Pd, and uses it for long-term and short-term memory network model adjustment.

[0013] Preferably, the historical data acquisition module includes a data collection unit and a data cleaning and standardization unit;

[0014] The data collection unit collects order data, material usage records and production cycle data through production management software and order records; material usage records include material consumption Qc and production quantity QL; production cycle data includes material usage time T and production time T c ;

[0015] The production batch fluctuation coefficient Bp and the material cycle peak intensity Mp are collected through statistics and analysis of historical production data; the production batch fluctuation coefficient Bp is obtained by the ratio of the standard deviation σp of each batch to the average output μp of all batches; the material cycle peak intensity Mp is obtained by the ratio of the maximum material usage Mmax to the average material usage μM;

[0016] The standardization unit cleans and standardizes the collected data to obtain a data set W;

[0017] Among them, cleaning includes using filters to remove outliers and process missing value data; standardization includes using the Z-score standard method to process the data.

[0018] Preferably, the production progress monitoring module includes an equipment operation status and production batch monitoring unit and a production rate calculation and instant progress index generation unit;

[0019] The equipment operation status and production batch monitoring unit is responsible for real-time monitoring of the equipment operation status, including the normal operation, failure and downtime of the equipment, and recording the completion progress of the production batch, including the completed output and production time T of each batch. c ;

[0020] The production rate calculation and instant progress index generation unit calculates the hourly production rate Ri according to the equipment operation status and the production batch completion progress, obtains the instant production progress index Pi, and compares it with the preset progress index PT=100 to obtain the completion progress of the current production;

[0021] The production rate Ri is calculated by the ratio of the completed output to the time taken:

[0022]

[0023] In the formula, QLt represents the production quantity completed at the current time point, and Te represents the elapsed time from the start of production to the current moment;

[0024] The instant production progress index Pi is obtained according to the ratio of the production rate Ri to the standard production rate Rtar, and is used to reflect the deviation of the production progress;

[0025]

[0026] Where Rtar represents the standard production rate, that is, the predetermined hourly output.

[0027] The completion progress of the current production is obtained by matching in the following way:

[0028] When the immediate production progress index Pi ≥ the preset progress index PT, it means that the production progress reaches the target progress and maintains the current production rate Ri;

[0029] When the immediate production progress index Pi is less than the preset progress index PT, it means that the production progress has not reached the target progress, and the current production rate Ri is adjusted.

[0030] Preferably, the consumption analysis and feature extraction module includes a consumption rate calculation and trend analysis unit and a consumption anomaly detection and fluctuation analysis unit;

[0031] The consumption rate calculation and trend analysis unit is responsible for extracting historical data of production and material consumption from the data set W and calculating the average consumption rate mc per unit time:

[0032] Specifically, by analyzing the relationship between the material consumption of each production batch and the corresponding production time, the consumption rate R is calculated. c,i , the consumption rate R c,i Obtained by the following formula:

[0033]

[0034] In the formula, Q c,i represents the material consumption of the i-th batch, T c,i represents the production time of the i-th batch;

[0035] By consumption rate R c,i Calculate the average consumption rate mc:

[0036]

[0037] In the formula, n represents the number of total batches;

[0038] By using the consumption rate R c,i The series is used as the basic data for analysis, seasonal and cyclical patterns are observed within a fixed period, and exponential smoothing is used to capture short-term changes in consumption rates and seasonal trends;

[0039] The smoothing formula is:

[0040] S t =α*R c,t +(1-α)*S t-1 ;

[0041] In the formula, S t represents the smoothed value at time t, S t-1 represents the smoothed value at time t-1, R c,t represents the actual consumption rate at time t, α represents the smoothing coefficient, and 0<α<1.

[0042] Preferably, the consumption anomaly detection and fluctuation analysis unit is responsible for analyzing the abnormal fluctuations in the material consumption process, based on the consumption rate R c,i The consumption abnormality coefficient ae is calculated and compared with the preset abnormality threshold coefficient Ta to determine the state of the production process;

[0043] The consumption abnormality coefficient ae is obtained by the following formula:

[0044]

[0045] In the formula, σ mc Represents the consumption rate R c,i The standard deviation of

[0046] The status of the production process is obtained by matching:

[0047] When the consumption abnormality coefficient ae>abnormal threshold coefficient Ta, it means that the fluctuation of material consumption is abnormal and the production process status is abnormal; the production efficiency is low or the equipment fails, and adjustment is required.

[0048] When the consumption abnormality coefficient ae ≤ the abnormal threshold coefficient Ta, it means that the fluctuation of material consumption is normal and the production process status is normal.

[0049] Preferably, the material demand forecasting module includes a feature construction and data preparation unit and a long short-term memory network model training and forecasting unit;

[0050] The feature construction and data preparation unit is responsible for constructing the features of the average consumption rate mc, the consumption abnormality coefficient ae, the production progress index Pi, the production batch fluctuation coefficient Bp and the material cycle peak intensity Mp, and integrating them into a feature vector X;

[0051] Standardize and normalize the feature vector X, and perform dimensionless operations to normalize the data to the same scale range;

[0052] The feature vector X is obtained in the following way:

[0053] X = {mc,ae,Pi,Bp,Mp};

[0054] The long short-term memory network model training and prediction unit trains the feature vector X, generates a material demand prediction model, and outputs the predicted future material demand Dfu;

[0055] The LSTM network model has an input layer, a hidden layer, and an output layer to capture long-term and short-term dependencies in the time series;

[0056] The prediction formula of the long short-term memory network model is:

[0057] ht=LSTM(Xt,h t-1 ,c t-1 );

[0058] Dfu=f(ht);

[0059] In the formula, ht represents the hidden state of the current time step, Xt represents the input feature vector X at time step t, and h t-1 represents the hidden state of the previous time step, c t-1 Represents the state of the memory unit at the previous time step, and f() represents the activation function of the output layer. It is used to convert the hidden state of the long short-term memory network model into the final prediction value.

[0060] Preferably, the purchase scheduling and inventory management module includes a purchase plan generation unit and an inventory strategy adjustment and automation instruction generation unit;

[0061] The purchasing plan generating unit calculates the required purchasing quantity Qor according to the future material demand quantity Dfu and the current inventory quantity Scu; and generates a purchasing plan according to the safety inventory quantity Ssa and the supplier delivery cycle Tle;

[0062] The purchase quantity Qor is obtained by the following formula:

[0063] Qor=Dfu-Scu+Ssa;

[0064] The procurement plan is specifically as follows:

[0065] Determine the purchase quantity Qor, generate a purchase order based on the purchase quantity Qor, set the purchase time, and deliver the materials on time according to the supplier's delivery cycle Tle, including the type and quantity of the materials and the selection of suppliers;

[0066] The inventory strategy adjustment and automated instruction generation unit updates the inventory strategy according to the procurement plan; and generates automated procurement instructions and sends them to management personnel to trigger procurement and material transfer;

[0067] The inventory strategy is specifically:

[0068] The current inventory level Scu is maintained between {Smin, Smax}; Smin represents the inventory valley value, and Smax represents the inventory peak value;

[0069] Adjust the inventory replenishment strategy and set the replenishment point Rop;

[0070] The replenishment point Rop is obtained by the following formula:

[0071] Rop = Dda*Tle+Ssa;

[0072] In the formula, Dda represents the average daily material demand, Tle represents the supplier delivery cycle, and Ssa represents the safety stock.

[0073] Preferably, the feedback and optimization module includes a deviation assessment and data acquisition unit and a model optimization and parameter adjustment unit;

[0074] The deviation assessment and data collection unit is responsible for collecting production and procurement data, including the consumption rate R c,i , purchase quantity Qor and inventory quantity Scu, and compare them with future material demand Dfu to obtain the forecast deviation coefficient Pd;

[0075] The prediction deviation coefficient Pd is obtained by the following formula:

[0076]

[0077] In the formula, T represents the time of material use, Dfu represents the future material demand, and Dac represents the actual material consumption.

[0078] After each round of material demand forecasting is executed, the actual material consumption data and inventory changes are collected; then, the deviation between the actual material consumption Dac and the future material demand Dfu is calculated, and the forecast deviation coefficient Pd is calculated.

[0079] Preferably, the model optimization and parameter adjustment unit determines the production status of the rope according to the predicted deviation coefficient Pd and the preset deviation threshold TPd;

[0080] The rope production status is obtained by the following matching method:

[0081] When the prediction deviation coefficient Pd ≤ the preset deviation threshold TPd, the parameters of the long short-term memory network model are not adjusted;

[0082] When the prediction deviation coefficient Pd> the preset deviation threshold TPd, the parameters of the long short-term memory network model are adjusted; the model is optimized to improve the accuracy of the next prediction; the optimization process includes adjusting the model's learning rate, number of hidden layer units, time step and other hyperparameters.

[0083] The prediction formula of the adjusted long short-term memory network model is:

[0084] ht=LSTM(Xt,h t-1 ,c t-1 ;θ);

[0085] Where θ represents the updated LSTM model parameters, including weight values ​​and bias terms.

[0086] A method for scheduling the production of ship cables comprises the following steps:

[0087] Step 1: The historical data collection module collects past order data, material usage records, production cycle data, production batch fluctuation coefficient Bp and material cycle peak intensity Mp, cleans and standardizes them, and integrates them into a data set W;

[0088] Step 2: The production progress monitoring module monitors the current production progress in real time according to the data set W, including the equipment operation status and the production batch completion progress, and records the hourly production rate changes to obtain the real-time production progress index Pi;

[0089] Step 3: The consumption analysis and feature extraction module analyzes the current production progress and the production progress index Pi, and extracts consumption parameters, including the average consumption rate mc and the consumption abnormality coefficient ae;

[0090] Step 4: The material demand forecasting module predicts future material demand through the long short-term memory network model based on the acquired consumption parameters to obtain the future material demand Dfu;

[0091] Step 5: The procurement scheduling and inventory management module automatically adjusts the material procurement plan and inventory strategy according to the future material demand Dfu;

[0092] Step 6: After the material procurement plan and inventory strategy are executed, the feedback and optimization module collects production and procurement data, evaluates the forecast accuracy, obtains the forecast deviation coefficient Pd, and uses it to adjust the long short-term memory network model.

[0093] The present invention provides a ship cable production scheduling method and scheduling system, which has the following beneficial effects:

[0094] (1) When the system is running, the production progress monitoring module can accurately grasp the hourly production rate changes and generate an instant production progress index Pi by monitoring the equipment status and production batch progress in real time. This data supports real-time adjustment of production plans to ensure more reasonable resource allocation and time scheduling during the production process, thereby improving production efficiency and reducing delays. At the same time, the system can predict future production trends based on changes in the production progress index, which helps to adjust production strategies and avoid production bottlenecks or shutdowns. The consumption analysis and feature extraction module can extract consumption parameters based on the current production progress and production progress index to accurately monitor material consumption. By analyzing and extracting consumption features, the system can identify whether there is abnormal material consumption in the production process, thereby discovering potential production problems in advance, such as equipment failures or abnormal production processes, and reducing waste and unnecessary resource consumption.

[0095] (2) By analyzing real-time data, the system can identify abnormal fluctuations in material usage during the production process and adjust material procurement and inventory management strategies in advance to ensure that there will be no shortage or surplus of materials during the production process. This combination of prediction and dynamic adjustment significantly improves material utilization and the response speed of the supply chain. By combining historical data collection, production progress monitoring and material demand forecasting, this system can achieve more intelligent procurement scheduling and inventory management. The system can automatically adjust material procurement plans and inventory strategies based on real-time production progress changes to avoid inventory backlogs caused by excess materials or production interruptions caused by material shortages. Through precise production scheduling, the system effectively reduces inventory costs, improves resource utilization efficiency, and ensures an efficient and smooth production process.

[0096] (3) The consumption analysis and feature extraction module can calculate the average consumption rate mc per unit time and analyze the seasonal and cyclical changes in material consumption through the consumption rate calculation and trend analysis unit. By analyzing the relationship between the material consumption of each production batch and the corresponding production time, and using the exponential smoothing method to capture the short-term changes and seasonal trends in the consumption rate, the system can more accurately predict future material demand and make scheduling decisions in advance. Through the consumption anomaly detection and fluctuation analysis unit, the system can monitor the fluctuations in material consumption during the production process in real time. When the consumption rate is abnormal, the system can determine whether the production process is in an abnormal state by calculating the consumption anomaly coefficient ae and comparing it with the preset anomaly threshold Ta.

[0097] (4) By integrating historical data analysis, production monitoring, material consumption analysis, demand forecasting and procurement scheduling, the entire production and supply chain process can be seamlessly connected. This highly integrated scheduling system can effectively coordinate the work of each link and improve the collaborative efficiency of production line and supply chain management. Through automated scheduling and feedback mechanisms, the system can quickly respond to changes in production demand, optimize resource allocation, reduce the risk of production stagnation, and thus improve overall production efficiency and supply chain stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0098] Figure 1 This is a schematic diagram of a block diagram of a ship cable production scheduling system of the present invention;

[0099] Figure 2 The present invention is a schematic diagram of the steps of a method for scheduling production of ship cables. DETAILED DESCRIPTION

[0100] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0101] Example 1

[0102] The present invention provides a ship rope production scheduling method and scheduling system, please refer to Figure 1 , including historical data collection module, production progress monitoring module, consumption analysis and feature extraction module, material demand forecasting module, procurement scheduling and inventory management module and feedback and optimization module;

[0103] The historical data collection module collects past order data, material usage records, production cycle data, production batch fluctuation coefficient Bp and material cycle peak intensity Mp, and performs cleaning and standardization processing to integrate them into a data set W;

[0104] The production progress monitoring module monitors the current production progress in real time according to the data set W, including the equipment operation status and the production batch completion progress, and records the hourly production rate changes to obtain the real-time production progress index Pi;

[0105] The consumption analysis and feature extraction module analyzes the current production progress and the production progress index Pi, and extracts consumption parameters, including the average consumption rate mc and the consumption abnormality coefficient ae;

[0106] The material demand forecasting module predicts future material demand through a long short-term memory network model based on the acquired consumption parameters to obtain future material demand Dfu;

[0107] The purchasing scheduling and inventory management module automatically adjusts the material purchasing plan and inventory strategy according to the future material demand Dfu;

[0108] After the material procurement plan and inventory strategy are executed, the feedback and optimization module collects production and procurement data, evaluates the prediction accuracy, obtains the prediction deviation coefficient Pd, and uses it for long-term and short-term memory network model adjustment.

[0109] In this embodiment, by using big data analysis and machine learning technology, especially by using the long short-term memory network model to predict material demand, the system can dynamically predict future material demand based on historical order data, production progress and material consumption. This dynamic prediction mechanism can respond to changes in production in real time, avoiding the limitations of traditional static prediction models, ensuring the accuracy and timeliness of material supply during the production process, and reducing the risk of material shortages or over-purchases.

[0110] The production progress monitoring module can accurately grasp the hourly production rate changes and generate an instant production progress index Pi by monitoring the equipment status and production batch progress in real time. This data supports real-time adjustment of production plans to ensure more reasonable resource allocation and time scheduling during the production process, thereby improving production efficiency and reducing delays. At the same time, the system can predict future production trends based on changes in the production progress index, which helps to adjust production strategies and avoid production bottlenecks or shutdowns. The consumption analysis and feature extraction module can extract consumption parameters based on the current production progress and production progress index to accurately monitor material consumption. By analyzing and extracting consumption features, the system can identify whether there is abnormal material consumption in the production process, thereby discovering potential production problems in advance, such as equipment failures or abnormal production processes, and reducing waste and unnecessary resource consumption.

[0111] The material demand forecasting module is based on accurate consumption forecasts, which can provide a scientific basis for the procurement scheduling and inventory management modules, and realize the automatic adjustment of material procurement plans and inventory strategies. This precise procurement and inventory management can reduce inventory backlogs and inventory costs, while avoiding production delays or interruptions caused by material shortages. The system ensures that materials are in place in a timely manner through automated procurement and transfer instructions, improving the flexibility and responsiveness of overall production. The adaptive feedback mechanism can automatically adjust the forecasting model when the forecast deviation is too large, improve the accuracy of the next forecast, and ensure that the forecast accuracy continues to improve over time. This innovation can cope with the ever-changing demands and uncertainties in actual production, and further improve the reliability and accuracy of the entire production scheduling system.

[0112] Example 2

[0113] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the historical data acquisition module includes a data collection unit and a data cleaning and standardization unit;

[0114] The data collection unit collects order data, material usage records and production cycle data through production management software and order records; material usage records include material consumption Qc and production quantity QL; production cycle data includes material usage time T and production time T c ;

[0115] The production batch fluctuation coefficient Bp and the material cycle peak intensity Mp are collected through statistics and analysis of historical production data; the production batch fluctuation coefficient Bp is obtained by the ratio of the standard deviation σp of each batch to the average output μp of all batches; the material cycle peak intensity Mp is obtained by the ratio of the maximum material usage Mmax to the average material usage μM;

[0116] The standardization unit cleans and standardizes the collected data to obtain a data set W;

[0117] Among them, cleaning includes using filters to remove outliers and process missing value data; standardization includes using the Z-score standard method to process the data.

[0118] The production progress monitoring module includes an equipment operation status and production batch monitoring unit and a production rate calculation and instant progress index generation unit;

[0119] The equipment operation status and production batch monitoring unit is responsible for real-time monitoring of the equipment operation status, including the normal operation, failure and downtime of the equipment, and recording the completion progress of the production batch, including the completed output and production time T of each batch.c ;

[0120] The production rate calculation and instant progress index generation unit calculates the hourly production rate Ri according to the equipment operation status and the production batch completion progress, obtains the instant production progress index Pi, and compares it with the preset progress index PT=100 to obtain the completion progress of the current production;

[0121] The production rate Ri is calculated by the ratio of the completed output to the time taken:

[0122]

[0123] In the formula, QLt represents the production quantity completed at the current time point, and Te represents the elapsed time from the start of production to the current moment;

[0124] The instant production progress index Pi is obtained according to the ratio of the production rate Ri to the standard production rate Rtar, and is used to reflect the deviation of the production progress;

[0125]

[0126] The completion progress of the current production is obtained by matching in the following way:

[0127] When the immediate production progress index Pi ≥ the preset progress index PT, it means that the production progress reaches the target progress and maintains the current production rate Ri;

[0128] When the immediate production progress index Pi is less than the preset progress index PT, it means that the production progress has not reached the target progress, and the current production rate Ri is adjusted.

[0129] In this embodiment, through the data collection unit and the data cleaning and standardization unit in the historical data acquisition module, order data, material usage records and production cycle data can be comprehensively collected, and these data can be cleaned and standardized. Through the accurate calculation of the production batch fluctuation coefficient Bp and the material cycle peak intensity Mp, the system can better understand the historical production pattern and identify potential problems in the production process. This standardized data processing method not only improves the quality of the data, but also ensures that the input data used by the prediction model is more accurate and consistent, providing reliable basic data for subsequent material demand forecasting, production progress monitoring and scheduling.

[0130] Through the implementation of the production progress monitoring module, especially the functions of the equipment operation status and production batch monitoring unit and the production rate calculation and instant progress index generation unit, the system can monitor the operation status of the equipment in real time and track the completion progress of the production batch. The generation of the instant production progress index Pi and the comparison with the preset progress index PT = 100 enable the production progress to be evaluated in real time and the production rate Ri to be adjusted according to the evaluation results.

[0131] The key innovation of the production progress monitoring module is that it can be adaptively adjusted according to the ratio of the immediate production progress index Pi to the standard production rate Rtar. When the production progress does not reach the target, the system can automatically adjust the production rate Ri to speed up the production rhythm to catch up with the progress. This dynamic adjustment mechanism effectively solves the problem of production progress lag in traditional production scheduling, making production scheduling more flexible and accurate, and can automatically respond to different production situations, improving the reliability and flexibility of production plans.

[0132] Example 3

[0133] This embodiment is explained in Example 2. Please refer to Figure 1 ,Specifically: the consumption analysis and feature extraction module includes a consumption rate calculation and trend analysis unit and a consumption anomaly detection and fluctuation analysis unit;

[0134] The consumption rate calculation and trend analysis unit is responsible for extracting historical data of production and material consumption from the data set W and calculating the average consumption rate mc per unit time:

[0135] Specifically, by analyzing the relationship between the material consumption of each production batch and the corresponding production time, the consumption rate R is calculated. c,i , the consumption rate R c,i Obtained by the following formula:

[0136]

[0137] In the formula, Q c,i represents the material consumption of the i-th batch, T c,i represents the production time of the i-th batch;

[0138] By consumption rate R c,i Calculate the average consumption rate mc:

[0139]

[0140] In the formula, n represents the number of total batches;

[0141] By using the consumption rate R c,iThe series is used as the basic data for analysis, seasonal and cyclical patterns are observed within a fixed period, and exponential smoothing is used to capture short-term changes in consumption rates and seasonal trends;

[0142] The smoothing formula is:

[0143] S t =α*R c,t +(1-α)*S t-1 ;

[0144] In the formula, S t represents the smoothed value at time t, S t-1 represents the smoothed value at time t-1, R c,t represents the actual consumption rate at time t, α represents the smoothing coefficient, and 0<α<1.

[0145] The consumption anomaly detection and fluctuation analysis unit is responsible for analyzing the abnormal fluctuations in the material consumption process based on the consumption rate R c,i The consumption abnormality coefficient ae is calculated and compared with the preset abnormality threshold coefficient Ta to determine the state of the production process;

[0146] The consumption abnormality coefficient ae is obtained by the following formula:

[0147]

[0148] In the formula, σ mc Represents the consumption rate R c,i The standard deviation of

[0149] The status of the production process is obtained by matching:

[0150] When the consumption abnormal coefficient ae>abnormal threshold coefficient Ta, it means that the fluctuation of material consumption is abnormal and the production process status is abnormal;

[0151] When the consumption abnormality coefficient ae ≤ the abnormal threshold coefficient Ta, it means that the fluctuation of material consumption is normal and the production process status is normal.

[0152] The material demand forecasting module includes a feature construction and data preparation unit and a long short-term memory network model training and forecasting unit;

[0153] The feature construction and data preparation unit is responsible for constructing the features of the average consumption rate mc, the consumption abnormality coefficient ae, the production progress index Pi, the production batch fluctuation coefficient Bp and the material cycle peak intensity Mp, and integrating them into a feature vector X;

[0154] Standardize and normalize the feature vector X, and perform dimensionless operations to normalize the data to the same scale range;

[0155] The feature vector X is obtained in the following way:

[0156] X = {mc,ae,Pi,Bp,Mp};

[0157] The long short-term memory network model training and prediction unit trains the feature vector X, generates a material demand prediction model, and outputs the predicted future material demand Dfu;

[0158] The LSTM network model has an input layer, a hidden layer, and an output layer to capture long-term and short-term dependencies in the time series;

[0159] The prediction formula of the long short-term memory network model is:

[0160] ht=LSTM(Xt,h t-1 ,c t-1 );

[0161] Dfu=f(ht);

[0162] In the formula, ht represents the hidden state of the current time step, Xt represents the input feature vector X at time step t, and h t-1 represents the hidden state of the previous time step, c t-1 represents the state of the memory unit at the previous time step, and f() represents the activation function of the output layer.

[0163] In this embodiment, the consumption analysis and feature extraction module can calculate the average consumption rate mc per unit time and analyze the seasonal and cyclical changes in material consumption through the consumption rate calculation and trend analysis unit. By analyzing the relationship between the material consumption of each production batch and the corresponding production time, and using the exponential smoothing method to capture the short-term changes and seasonal trends in the consumption rate, the system can more accurately predict future material requirements and make scheduling decisions in advance. Through the consumption anomaly detection and fluctuation analysis unit, the system can monitor the fluctuations in material consumption during the production process in real time. When the consumption rate is abnormal, the system can determine whether the production process is in an abnormal state by calculating the consumption anomaly coefficient ae and comparing it with the preset abnormal threshold Ta.

[0164] Through accurate prediction, the system can adjust material procurement and inventory strategies in advance to ensure that materials are in place in time and avoid production interruptions or material waste. Through accurate material demand forecasting, anomaly detection and fluctuation analysis, production progress monitoring and other functions, the production scheduling and material management processes are optimized. The system can automatically adjust production strategies and inventory management based on real-time monitoring of production progress and material consumption, reduce the risk of excessive inventory and material shortages, and improve production efficiency and material utilization. This comprehensive optimization solution not only improves the flexibility of the production process, but also effectively improves the response speed and resource utilization of the entire supply chain, providing enterprises with more efficient and accurate production scheduling management capabilities.

[0165] Example 4

[0166] This embodiment is explained in Example 3, please refer to Figure 1 ,Specifically: the procurement scheduling and inventory management module includes a procurement plan generating unit and an inventory strategy adjustment and automated instruction generating unit;

[0167] The purchasing plan generating unit calculates the required purchasing quantity Qor according to the future material demand quantity Dfu and the current inventory quantity Scu; and generates a purchasing plan according to the safety inventory quantity Ssa and the supplier delivery cycle Tle;

[0168] The purchase quantity Qor is obtained by the following formula:

[0169] Qor=Dfu-Scu+Ssa;

[0170] The procurement plan is specifically as follows:

[0171] Determine the purchase quantity Qor, generate a purchase order based on the purchase quantity Qor, set the purchase time, and deliver the materials on time according to the supplier's delivery cycle Tle, including the type and quantity of the materials and the selection of suppliers;

[0172] The inventory strategy adjustment and automated instruction generation unit updates the inventory strategy according to the procurement plan; and generates automated procurement instructions and sends them to management personnel to trigger procurement and material transfer;

[0173] The inventory strategy is specifically:

[0174] The current inventory level Scu is maintained between {Smin, Smax}; Smin represents the inventory valley value, and Smax represents the inventory peak value;

[0175] Adjust the inventory replenishment strategy and set the replenishment point Rop;

[0176] The replenishment point Rop is obtained by the following formula:

[0177] Rop = Dda*Tle+Ssa;

[0178] In the formula, Dda represents the average daily material demand, Tle represents the supplier delivery cycle, and Ssa represents the safety stock.

[0179] The feedback and optimization module includes a deviation assessment and data acquisition unit and a model optimization and parameter adjustment unit;

[0180] The deviation assessment and data collection unit is responsible for collecting production and procurement data, including the consumption rate R c,i , purchase quantity Qor and inventory quantity Scu, and compare them with future material demand Dfu to obtain the forecast deviation coefficient Pd;

[0181] The prediction deviation coefficient Pd is obtained by the following formula:

[0182]

[0183] In the formula, T represents the time of material use, Dfu represents the future material demand, and Dac represents the actual material consumption.

[0184] The model optimization and parameter adjustment unit determines the production status of the rope according to the predicted deviation coefficient Pd and the preset deviation threshold TPd;

[0185] The rope production status is obtained by the following matching method:

[0186] When the prediction deviation coefficient Pd ≤ the preset deviation threshold TPd, the parameters of the long short-term memory network model are not adjusted;

[0187] When the prediction deviation coefficient Pd> the preset deviation threshold TPd, the parameters of the long short-term memory network model are adjusted;

[0188] The prediction formula of the adjusted long short-term memory network model is:

[0189] ht=LSTM(Xt,h t-1 ,c t-1 ;θ);

[0190] Where θ represents the updated LSTM model parameters.

[0191] In this embodiment, the procurement scheduling and inventory management module accurately predicts and generates the required procurement quantity Qor by calculating the future material demand Dfu and the current inventory Scu. Through the safety inventory Ssa and the supplier delivery cycle Tle, the generated procurement plan can ensure that the materials are in place in time before the peak demand period, avoiding production interruptions due to material shortages. This procurement plan generation mechanism effectively reduces production delays caused by insufficient material supply during the production process and improves production continuity and efficiency.

[0192] The inventory strategy adjustment and automated instruction generation unit can automatically adjust inventory strategies and generate automated procurement instructions based on future material demand forecasts and procurement plans, and send them to management personnel to trigger procurement and material transfers. The intelligent setting of replenishment points enables the system to flexibly respond to uncertainties in production and avoid inventory shortages or excesses caused by sudden demand or production interruptions. This dynamic replenishment strategy based on demand and delivery cycle can further improve material flow efficiency and optimize inventory management.

[0193] If the forecast deviation coefficient exceeds the preset threshold TPd, the system will automatically adjust the parameters of the long short-term memory network model to optimize the forecast results. Through this mechanism, the system can continuously optimize the forecast accuracy and adjust the material demand forecast model to make it more accurate in the face of changing production environments and demand fluctuations, thereby reducing errors in procurement and inventory management, and further improving the response speed and accuracy of the supply chain.

[0194] The automated procurement and scheduling system reduces the need for manual intervention and reduces scheduling problems caused by human errors. The system provides decision support through big data analysis and intelligent algorithms to help managers make more accurate procurement decisions and inventory adjustments. The automated process ensures the continuity of material supply, avoids production delays caused by manual decision-making lags or misjudgments, and improves the efficiency and reliability of the overall supply chain.

[0195] Example 5

[0196] A method for scheduling the production of ship ropes, please refer to Figure 2 , specifically: including the following steps:

[0197] Step 1: The historical data collection module collects past order data, material usage records, production cycle data, production batch fluctuation coefficient Bp and material cycle peak intensity Mp, cleans and standardizes them, and integrates them into a data set W;

[0198] Step 2: The production progress monitoring module monitors the current production progress in real time according to the data set W, including the equipment operation status and the production batch completion progress, and records the hourly production rate changes to obtain the real-time production progress index Pi;

[0199] Step 3: The consumption analysis and feature extraction module analyzes the current production progress and the production progress index Pi, and extracts consumption parameters, including the average consumption rate mc and the consumption abnormality coefficient ae;

[0200] Step 4: The material demand forecasting module predicts future material demand through the long short-term memory network model based on the acquired consumption parameters to obtain the future material demand Dfu;

[0201] Step 5: The procurement scheduling and inventory management module automatically adjusts the material procurement plan and inventory strategy according to the future material demand Dfu;

[0202] Step 6: After the material procurement plan and inventory strategy are executed, the feedback and optimization module collects production and procurement data, evaluates the forecast accuracy, obtains the forecast deviation coefficient Pd, and uses it to adjust the long short-term memory network model.

[0203] In this embodiment, the historical data collection module collects past order data, material usage records, production cycle data, and key parameters including the production batch fluctuation coefficient Bp and the material cycle peak intensity Mp, and cleans and standardizes them to form a data set W, which provides accurate data support for subsequent production scheduling. The standardized processing of the data set ensures the consistency and accuracy of the data, helps to eliminate the noise in the historical data, and makes subsequent predictions and optimizations more accurate. Through a data-driven approach, the interference of human factors in the production process can be effectively reduced, and the accuracy and reliability of production scheduling can be improved.

[0204] The production progress monitoring module monitors the equipment operation status and production batch progress in real time, generates an instant production progress index Pi, and compares it with the preset progress target. This mechanism can promptly detect production progress deviations and quickly take corresponding measures to adjust them, ensuring that the production progress always remains within a reasonable range and avoiding production interruptions or delays caused by equipment failures or production delays. Through the application of this module, the visualization and dynamic adjustment of production progress greatly enhance the flexibility and emergency response capabilities of the production process.

[0205] The feedback and optimization module provides real-time feedback on the execution effect of material procurement plans and inventory strategies, evaluates the accuracy of forecasts based on production and procurement data, and calculates the forecast deviation coefficient Pd. When the forecast deviation is large, the system can automatically adjust the parameters of the long short-term memory network model to optimize the forecast results. This feedback-based continuous optimization mechanism enables the system to continuously learn and improve based on real-time data, gradually improve forecast accuracy and scheduling efficiency, and enhance the system's adaptability in different production scenarios. By integrating historical data analysis, production monitoring, material consumption analysis, demand forecasting and procurement scheduling, the entire production and supply chain process can be seamlessly connected. This highly integrated scheduling system can effectively coordinate the work of each link and improve the collaborative efficiency of production line and supply chain management. Through automated scheduling and feedback mechanisms, the system can quickly respond to changes in production demand, optimize resource allocation, reduce the risk of production stagnation, and thus improve overall production efficiency and supply chain stability.

[0206] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A ship rope production scheduling system, characterized by: It includes historical data collection module, production progress monitoring module, consumption analysis and feature extraction module, material demand forecasting module, procurement scheduling and inventory management module and feedback and optimization module; The historical data collection module collects past order data, material usage records, production cycle data, production batch fluctuation coefficient Bp and material cycle peak intensity Mp, and performs cleaning and standardization processing to integrate them into a data set W; The production progress monitoring module monitors the current production progress in real time according to the data set W, including the equipment operation status and the production batch completion progress, and records the hourly production rate changes to obtain the real-time production progress index Pi; The consumption analysis and feature extraction module analyzes the current production progress and the production progress index Pi, and extracts consumption parameters, including the average consumption rate mc and the consumption abnormality coefficient ae; The material demand forecasting module predicts future material demand through a long short-term memory network model based on the acquired consumption parameters to obtain future material demand Dfu; The purchasing scheduling and inventory management module automatically adjusts the material purchasing plan and inventory strategy according to the future material demand Dfu; After the material procurement plan and inventory strategy are executed, the feedback and optimization module collects production and procurement data, evaluates the prediction accuracy, obtains the prediction deviation coefficient Pd, and uses it for long-term and short-term memory network model adjustment.

2. A ship rope production scheduling system according to claim 1, characterized in that: The historical data acquisition module includes a data collection unit and a data cleaning and standardization unit; The data collection unit collects order data, material usage records and production cycle data through production management software and order records; material usage records include material consumption Qc and production quantity QL; production cycle data includes material usage time T and production time T c ; The production batch fluctuation coefficient Bp and the material cycle peak intensity Mp are collected through statistics and analysis of historical production data; the production batch fluctuation coefficient Bp is obtained by the ratio of the standard deviation σp of each batch to the average output μp of all batches; the material cycle peak intensity Mp is obtained by the ratio of the maximum material usage Mmax to the average material usage μM; The standardization unit cleans and standardizes the collected data to obtain a data set W; Among them, cleaning includes using filters to remove outliers and process missing value data; standardization includes using the Z-score standard method to process the data.

3. A ship rope production scheduling system according to claim 1, characterized in that: The production progress monitoring module includes an equipment operation status and production batch monitoring unit and a production rate calculation and instant progress index generation unit; The equipment operation status and production batch monitoring unit is responsible for real-time monitoring of the equipment operation status, including the normal operation, failure and downtime of the equipment, and recording the completion progress of the production batch, including the completed output and production time T of each batch. c ; The production rate calculation and instant progress index generation unit calculates the hourly production rate Ri according to the equipment operation status and the production batch completion progress, obtains the instant production progress index Pi, and compares it with the preset progress index PT=100 to obtain the completion progress of the current production; The production rate Ri is calculated by the ratio of the completed output to the time taken: In the formula, QLt represents the production quantity completed at the current time point, and Te represents the elapsed time from the start of production to the current moment; The instant production progress index Pi is obtained according to the ratio of the production rate Ri to the standard production rate Rtar, and is used to reflect the deviation of the production progress; The completion progress of the current production is obtained by matching in the following way: When the immediate production progress index Pi ≥ the preset progress index PT, it means that the production progress reaches the target progress and maintains the current production rate Ri; When the immediate production progress index Pi is less than the preset progress index PT, it means that the production progress has not reached the target progress, and the current production rate Ri is adjusted.

4. A ship rope production scheduling system according to claim 1, characterized in that: The consumption analysis and feature extraction module includes a consumption rate calculation and trend analysis unit and a consumption anomaly detection and fluctuation analysis unit; The consumption rate calculation and trend analysis unit is responsible for extracting historical data of production and material consumption from the data set W and calculating the average consumption rate mc per unit time: Specifically, by analyzing the relationship between the material consumption of each production batch and the corresponding production time, the consumption rate R is calculated. c,i , the consumption rate R c,i Obtained by the following formula: In the formula, Q c,i represents the material consumption of the i-th batch, T c,i represents the production time of the i-th batch; By consumption rate R c,i Calculate the average consumption rate mc: In the formula, n represents the number of total batches; By using the consumption rate R c,i The series is used as the basic data for analysis, seasonal and cyclical patterns are observed within a fixed period, and exponential smoothing is used to capture short-term changes in consumption rates and seasonal trends; The smoothing formula is: S t =α*R c,t +(1-a)*S t-1 ; In the formula, S t represents the smoothed value at time t, S t-1 represents the smoothed value at time t-1, R c,t represents the actual consumption rate at time t, α represents the smoothing coefficient, and 0<α<1.

5. A ship rope production scheduling system according to claim 4, characterized in that: The consumption anomaly detection and fluctuation analysis unit is responsible for analyzing the abnormal fluctuations in the material consumption process based on the consumption rate R c,i The consumption abnormality coefficient ae is calculated and compared with the preset abnormality threshold coefficient Ta to determine the state of the production process; The consumption abnormality coefficient ae is obtained by the following formula: In the formula, σ mc Represents the consumption rate R c,i The standard deviation of The status of the production process is obtained by matching: When the consumption abnormal coefficient ae>abnormal threshold coefficient Ta, it means that the fluctuation of material consumption is abnormal and the production process status is abnormal; When the consumption abnormality coefficient ae ≤ the abnormal threshold coefficient Ta, it means that the fluctuation of material consumption is normal and the production process status is normal.

6. A ship rope production scheduling system according to claim 1, characterized in that: The material demand forecasting module includes a feature construction and data preparation unit and a long short-term memory network model training and forecasting unit; The feature construction and data preparation unit is responsible for constructing the features of the average consumption rate mc, the consumption abnormality coefficient ae, the production progress index Pi, the production batch fluctuation coefficient Bp and the material cycle peak intensity Mp, and integrating them into a feature vector X; Standardize and normalize the feature vector X, and perform dimensionless operations to normalize the data to the same scale range; The feature vector X is obtained in the following way: X = {mc,ae,Pi,Bp,Mp}; The long short-term memory network model training and prediction unit trains the feature vector X, generates a material demand prediction model, and outputs the predicted future material demand Dfu; The LSTM network model has an input layer, a hidden layer, and an output layer to capture long-term and short-term dependencies in the time series; The prediction formula of the long short-term memory network model is: ht=LSTM(Xt,h t-1 ,c t-1 ); Dfu=f(ht); In the formula, ht represents the hidden state of the current time step, Xt represents the input feature vector X at time step t, and h t-1 represents the hidden state of the previous time step, c t-1 represents the state of the memory unit at the previous time step, and f() represents the activation function of the output layer.

7. A ship rope production scheduling system according to claim 1, characterized in that: The procurement scheduling and inventory management module includes a procurement plan generation unit and an inventory strategy adjustment and automation instruction generation unit; The purchasing plan generating unit calculates the required purchasing quantity Qor according to the future material demand quantity Dfu and the current inventory quantity Scu; and generates a purchasing plan according to the safety inventory quantity Ssa and the supplier delivery cycle Tle; The purchase quantity Qor is obtained by the following formula: Qor=Dfu-Scu+Ssa; The procurement plan is specifically as follows: Determine the purchase quantity Qor, generate a purchase order based on the purchase quantity Qor, set the purchase time, and deliver the materials on time according to the supplier's delivery cycle Tle, including the type and quantity of the materials and the selection of suppliers; The inventory strategy adjustment and automated instruction generation unit updates the inventory strategy according to the procurement plan; and generates automated procurement instructions and sends them to management personnel to trigger procurement and material transfer; The inventory strategy is specifically: The current inventory level Scu is maintained between {Smin, Smax}; Smin represents the valley value of the inventory level, and Smax represents the peak value of the inventory level; Adjust the inventory replenishment strategy and set the replenishment point Rop; The replenishment point Rop is obtained by the following formula: Rop = Dda*Tle+Ssa; In the formula, Dda represents the average daily material demand, Tle represents the supplier delivery cycle, and Ssa represents the safety stock.

8. A ship rope production scheduling system according to claim 1, characterized in that: The feedback and optimization module includes a deviation assessment and data acquisition unit and a model optimization and parameter adjustment unit; The deviation assessment and data collection unit is responsible for collecting production and procurement data, including the consumption rate R c,i , purchase quantity Qor and inventory quantity Scu, and compare them with future material demand Dfu to obtain the forecast deviation coefficient Pd; The prediction deviation coefficient Pd is obtained by the following formula: In the formula, T represents the time of material use, Dfu represents the future material demand, and Dac represents the actual material consumption.

9. A ship rope production scheduling system according to claim 8, characterized in that: The model optimization and parameter adjustment unit determines the production status of the rope according to the predicted deviation coefficient Pd and the preset deviation threshold TPd; The rope production status is obtained by the following matching method: When the prediction deviation coefficient Pd ≤ the preset deviation threshold TPd, the parameters of the long short-term memory network model are not adjusted; When the prediction deviation coefficient Pd> the preset deviation threshold TPd, the parameters of the long short-term memory network model are adjusted; The prediction formula of the adjusted long short-term memory network model is: ht=LSTM(Xt,h t-1 ,c t-1 ;θ); Where θ represents the updated LSTM model parameters.

10. A ship cable production scheduling method, applied to a ship cable production scheduling system as claimed in any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: The historical data collection module collects past order data, material usage records, production cycle data, production batch fluctuation coefficient Bp and material cycle peak intensity Mp, cleans and standardizes them, and integrates them into a data set W; Step 2: The production progress monitoring module monitors the current production progress in real time according to the data set W, including the equipment operation status and the production batch completion progress, and records the hourly production rate changes to obtain the real-time production progress index Pi; Step 3: The consumption analysis and feature extraction module analyzes the current production progress and the production progress index Pi, and extracts consumption parameters, including the average consumption rate mc and the consumption abnormality coefficient ae; Step 4: The material demand forecasting module predicts future material demand through the long short-term memory network model based on the acquired consumption parameters to obtain the future material demand Dfu; Step 5: The procurement scheduling and inventory management module automatically adjusts the material procurement plan and inventory strategy according to the future material demand Dfu; Step 6: After the material procurement plan and inventory strategy are executed, the feedback and optimization module collects production and procurement data, evaluates the forecast accuracy, obtains the forecast deviation coefficient Pd, and uses it to adjust the long short-term memory network model.

Images