Production scheduling strategy optimization method and system based on state awareness and user decision

Abstract

The invention discloses a production scheduling strategy optimization method and system based on state awareness and user decision. The method comprises the steps of collecting real-time production state data; performing standardization processing on the collected data, and converting the data into a unified state vector; calculating a dynamic weight adjustment amount of each scheduling rule according to the state vector; calculating the rule priority of each scheduling rule; smoothing the dynamic weight adjustment amount to obtain a final weight adjustment value; and executing a production scheduling strategy according to the rule priority and the final weight adjustment value, performing effect evaluation on an actual production effect, adjusting user decision model parameters, and performing adaptive optimization on the production scheduling strategy. According to the method, a dynamic weight adjustment mechanism based on the production state is established, the problem that weight parameters are fixed in traditional production scheduling is solved, and the adaptability and accuracy of a production scheduling strategy are improved.

Description

Technical Field

[0001] The present invention relates to the field of manufacturing industry information systems, and in particular to a production scheduling strategy optimization method and system based on state perception and user decision-making. Background Art

[0002] At present, the manufacturing industry mainly adopts hybrid algorithm solutions based on multiple constraints, including the combined application of genetic algorithms, simulated annealing algorithms and heuristic algorithms. The system performs comprehensive calculations and outputs scheduling solutions by setting the weights of multiple key indicators such as minimum delay, equipment utilization, shortest flow time, inventory cost, etc. However, the fixed weight mode in the existing technology lacks flexibility. Users need to frequently adjust parameters manually to adapt to the complex and changing production status, and usually rely on actual experience to adjust scheduling parameters. Summary of the invention

[0003] In order to solve the problems of low flexibility and strong dependence of scheduling schemes in the prior art, the present invention provides a production scheduling strategy optimization method and system based on state perception and user decision-making.

[0004] To this end, the technical solution adopted in the present invention is:

[0005] A production scheduling strategy optimization method based on state perception and user decision-making, the method comprising:

[0006] Collect real-time production status data;

[0007] Standardize the collected data and convert it into a unified state vector;

[0008] Calculate the dynamic weight adjustment amount of each dispatch rule according to the state vector;

[0009] Calculate the rule priority of each scheduling rule;

[0010] Smoothing the dynamic weight adjustment amount to obtain the final weight adjustment value;

[0011] Execute the production scheduling strategy according to the rule priority and the final weight adjustment value, evaluate the actual production effect, adjust the user decision model parameters and adaptively optimize the production scheduling strategy.

[0012] According to the above scheme, the real-time production status data includes: equipment status data, inventory status data and order status data.

[0013] According to the above scheme, the state vector is specifically calculated from the equipment status data, inventory status data and order status data; the equipment status data is obtained by standardizing the equipment utilization rate, failure rate and production capacity balance; the inventory status data is obtained by standardizing the inventory turnover rate, inventory pressure index and raw material supply risk; the order status data is obtained by standardizing the order delay rate, delivery urgency and plan completion rate.

[0014] According to the above scheme, the dynamic weight adjustment amount is specifically obtained by multiplying the basic adjustment coefficient, the rule response function, the time series attenuation function, the historical cumulative impact function, and the effect evaluation function.

[0015] According to the above scheme, the rule priority is specifically calculated by combining the urgency, importance and timeliness of each scheduling rule according to the weights.

[0016] According to the above scheme, the final weight adjustment value is specifically obtained by adding the current weight adjustment value to the product of the smoothing coefficient and the dynamic weight adjustment amount.

[0017] According to the above scheme, the calculation of the dynamic weight adjustment amount and the final weight adjustment value is subject to system constraints, and the system constraints include:

[0018] The weight adjustment value of each dispatch rule is between 0 and 1;

[0019] The sum of the weight adjustment values ​​of each dispatch rule is equal to 1;

[0020] The absolute value of the dynamic weight adjustment amount of each dispatching rule is less than or equal to the preset value;

[0021] The weight adjustment values ​​of each dispatching rule are subtracted from each other, and the absolute value of the difference is less than or equal to the preset value.

[0022] According to the above scheme, the user decision model is obtained in the following way:

[0023] Collect user historical decision data, including production status and adjustment strategy data;

[0024] Analyze user historical decision data, extract key features related to adjustment strategies, and conduct differentiated analysis of user adjustment behaviors;

[0025] Establish a mapping model based on production status and user adjustment behavior;

[0026] Calculate the similarity between different production status scenarios and identify historical scenarios similar to the current production status;

[0027] Further improve the mapping model and establish user decision models under different production conditions;

[0028] The adjusting of the user decision model parameters specifically refers to generating a recommended dynamic weight adjustment value based on the user decision model of the current production status.

[0029] According to the above scheme, the adaptive optimization of the production scheduling strategy is achieved by adjusting the dynamic weight adjustment amount of each scheduling rule and the rule priority to minimize the optimization objective function value.

[0030] A production scheduling strategy optimization system based on state perception and user decision-making, comprising:

[0031] Collection module, used to collect real-time production status data;

[0032] The state evaluation module is used to standardize the collected data and convert it into a unified state vector;

[0033] A dynamic weight calculation module is used to calculate the dynamic weight adjustment amount of each dispatching rule according to the state vector;

[0034] A priority evaluation module, used to calculate the rule priority of each scheduling rule;

[0035] The weight smoothing processing module is used to smooth the dynamic weight adjustment amount to obtain the final weight adjustment value;

[0036] The execution module is used to execute the production scheduling strategy according to the rule priority and the final weight adjustment value, evaluate the actual production effect, adjust the user decision model parameters and adaptively optimize the production scheduling strategy.

[0037] The beneficial effects produced by the present invention are:

[0038] 1. By establishing a dynamic weight adjustment mechanism based on production status, the problem of fixed weight parameters in traditional production scheduling is solved, and the adaptability and accuracy of the scheduling scheme are improved.

[0039] 2. By introducing a weight smoothing mechanism and a multi-dimensional effect evaluation model, the problem of unstable scheduling caused by the complex and changeable production environment can be solved to ensure the continuity and stability of system adjustments.

[0040] 3. By establishing a user decision-making learning model, the intelligent extraction and reuse of scheduling experience can be realized, the defects of strong reliance on experience and low efficiency in manual adjustment of scheduling parameters can be overcome, and intelligent parameter adjustment suggestions can be provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a schematic diagram of a method flow of an embodiment of the present invention;

[0042] Figure 2 It is a schematic diagram of the production scheduling strategy optimization system architecture according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0044] like Figure 1 As shown, the present invention provides a production scheduling strategy optimization method based on state perception and user decision-making, comprising:

[0045] S1. Collect real-time production status data.

[0046] The system continuously obtains and implements production status data, including equipment (E) status data, inventory (I) status data, and order (O) status data.

[0047] S2. Standardize the collected data and convert it into a unified state vector.

[0048] The collected production status data of different types are standardized and converted into a unified state vector S, which reflects the overall status of the current production system and provides a basis for subsequent decision-making. Through standardization, different types of indicators can eliminate dimensional differences and ensure the consistency and accuracy of subsequent analysis.

[0049] The state vector S is defined as:

[0050] S=[SE,SI,SO]

[0051] Where SE represents the equipment status data, SE = f1 (equipment utilization, failure rate, capacity balance);

[0052] SI represents inventory status data, SI = f2 (inventory turnover rate, inventory pressure index, raw material supply risk);

[0053] SO represents the order status data, SO=f3 (order delay rate, delivery urgency, plan completion rate).

[0054] Multiple indicators in the three dimensions of equipment (E), inventory (I), and order (O) are converted into standardized values ​​through functions f1, f2, and f3 to achieve unified measurement of indicators in different dimensions and dimensions, which is convenient for subsequent calculations.

[0055] S3. Calculate the dynamic weight adjustment of each scheduling rule based on the state vector.

[0056] After obtaining the state vector S, the system responds to changes in production status by calculating the dynamic weight adjustment ΔW of each scheduling rule. The system will comprehensively consider factors such as equipment operation, inventory status, order information, and combine historical adjustments and timeliness to calculate the adjustment amount. Dynamic weight adjustment can ensure the flexibility of production scheduling rules, so that the production system can maintain efficient operation when responding to changes.

[0057] Dynamic weight adjustment ΔW = α*R(S)*P(t)*C(h)*E(v)

[0058] Where: α is the basic adjustment coefficient (0<α≤1); R(S) is the rule response function, triggered by the state vector S; P(t) is the time series attenuation function, P(t) = e^(-λt), used to control the adjustment frequency; C(h) is the historical cumulative impact function, C(h) = 1 / (1+βh), h is the number of historical adjustments; E(v) is the effect evaluation function, based on the effect feedback of the last adjustment; λ, β are adjustable parameters.

[0059] S4. Calculate the rule priority of each scheduling rule.

[0060] For multiple possible scheduling rules, the system evaluates the priority of each rule by calculating factors such as the urgency, importance, and timeliness of the rule. This process ensures that key tasks can be executed first, avoiding resource conflicts and production bottlenecks. This step helps the system prioritize the most urgent and important production tasks when making decisions, and optimizes resource allocation.

[0061] Rule priority Priority=ω1*U+ω2*I+ω3*T

[0062] Among them: U represents the urgency of the rule, I represents the importance of the rule, T represents the timeliness factor, and U, I, and T are between 0 and 1; ω1, ω2, and ω3 are weight coefficients, and Σωi=1.

[0063] Combine urgency (U), importance (I), and timeliness (T) according to weights, assign processing priorities to rules, and resolve rule conflicts.

[0064] S5. Smoothing the dynamic weight adjustment amount to obtain a final weight adjustment value.

[0065] After determining the dynamic weight adjustment amount, the system uses a smoothing mechanism to avoid excessive weight changes. By controlling the smoothing coefficient, the system ensures that the amplitude of each adjustment remains within a reasonable range, thereby avoiding production process instability caused by excessive adjustment. The smoothing mechanism ensures that the system responds quickly to changes in production demand while ensuring overall stability and controllability.

[0066] Final weight adjustment value W(t+1)=W(t)+γ(t)*ΔW

[0067] Where W(t) represents the weight value at the current moment; ΔW represents the dynamic weight adjustment calculated in S3; γ(t) is the smoothing coefficient, Where k is an adjustable parameter used to control the adjustment range.

[0068] The actual adjustment range is controlled by the smoothing coefficient γ(t). The larger the adjustment amount, the stronger the smoothing effect. It can avoid drastic fluctuations in weights to maintain system stability.

[0069] S6. Execute the production scheduling strategy according to the rule priority and the final weight adjustment value, evaluate the actual production effect, adjust the user decision model parameters and adaptively optimize the production scheduling strategy.

[0070] After each weight adjustment, the system will provide real-time feedback by evaluating the actual production effect after the adjustment. This feedback mechanism ensures that the system can adjust the rule parameters and execution strategy according to the actual production effect, thereby further improving the accuracy and response speed of decision-making. The core role of the feedback adjustment mechanism is to enable the system to continuously correct and optimize the scheduling strategy in practice.

[0071] Effect evaluation function E(v) = (1 + tanh(k*(v-v0))) / 2

[0072] Among them, v represents the current performance indicator value; v0 represents the target threshold; k represents the sensitivity parameter.

[0073] The system records historical adjustments and their effects, and uses an adaptive optimization mechanism to continuously adjust parameters and priorities. The effects of each adjustment are fed back and used to adjust the user decision model parameters to optimize subsequent decisions. Through continuous adjustment and optimization, the system can achieve self-correction and improvement in a constantly changing production environment, ensuring long-term operational stability and efficiency.

[0074] The optimization objective function is constructed as follows:

[0075] F=Σ(αi*Pi+βi*Ei-γi*Ci)

[0076] Among them, Pi represents the performance improvement index; Ei represents the adjustment efficiency index; Ci represents the adjustment cost index; αi, βi, γi represent the weight coefficients of the corresponding parameters.

[0077] By minimizing the F value, the production scheduling strategy is adaptively optimized to obtain the optimal production scheduling strategy.

[0078] Among them, the method for establishing the user decision model is:

[0079] The system collects historical decision data of users, including production status, original scheduling strategy (weight parameter set), adjusted strategy (modified weight parameter), adjustment reason mark, and final adopted plan. It analyzes the data and extracts key features related to the adjustment strategy. It conducts differentiated analysis on user adjustment behaviors and obtains adjustment trigger features, such as key feature adjustment direction, amplitude, parameter combination change mode, and adjustment trigger feature key indicator overrun, resource conflict, and delivery risk.

[0080] For production status and user adjustment behavior, the system establishes a mapping model, taking the state vector of production status and user adjustment behavior as the input layer, state feature extraction, decision pattern recognition and correlation analysis as the hidden layer, and outputting weight adjustment suggestions and adjustment confidence;

[0081] The system calculates the similarity between different production scenarios, identifies historical scenarios similar to the current production status, and provides a reference for the current production status so that quick decisions can be made in similar situations and repeated calculations can be reduced. Sim(Sa,Sb)=exp(-||Sa-Sb||2 / σ2)

[0082] Among them, Sa and Sb represent two production scenarios, and σ controls the similarity decay rate. The core indicators of its calculation are equipment load status, order urgency, inventory pressure level, and are constrained by resource limitations, process requirements, and delivery requirements.

[0083] The system further analyzes the decision-making pattern of user preferences and establishes the decision preference model of users under different production conditions by extracting key decision features. P(ΔW|S)=Σwi*fi(S)

[0084] Where fi represents the decision characteristics, including historical adjustment frequency, adjustment amplitude distribution, parameter sensitivity, and effect evaluation; wi represents the weight coefficient, including time attenuation factor, effect feedback factor, and confidence factor.

[0085] Based on the user decision model of the current production status, the system generates an optimal dynamic weight adjustment strategy ΔW*, ΔW*=argmax P(ΔW|St), and provides users with an optimized adjustment plan to ensure the smooth progress of the production process.

[0086] Through the user decision model, the system can predict the user's decision-making behavior under different production conditions, and give the system-recommended dynamic weight adjustment strategy as an influencing factor for parameter adjustment in adaptive optimization to optimize the production scheduling strategy.

[0087] In addition, in order to ensure the stability and effectiveness of the system and the rational use of various resources, the dynamic weight adjustment amount and the final weight adjustment value are subject to certain system constraints. The main system constraints include:

[0088] (1) Weight range constraints

[0089] 0≤Wi≤1

[0090] Each weight Wi needs to be kept within a valid range, that is, the value of each weight must be between 0 and 1. This constraint ensures the reasonable distribution of weights for each scheduling rule, avoiding situations where the weight value is too large or too small, thereby maintaining the balance of the scheduling system.

[0091] (2) Weights and Constraints

[0092] ΣWi=1

[0093] The sum of the weights of all rules must be 1. This constraint ensures that the system does not have an imbalance in resource allocation when adjusting weights. By forcing the sum of weights to be 1, the system can maintain a relative balance between the rules during adjustments and avoid having a single rule with too large or too small a weight.

[0094] (3) Weight change rate constraint

[0095] |ΔWi|≤Δmax

[0096] In each weight adjustment, the system needs to limit the maximum adjustment range of each weight to Δmax. This constraint ensures that the adjustment will not be too drastic and avoids system instability caused by excessive adjustment. The weight change rate constraint ensures that the system can respond smoothly to changes in production status.

[0097] (4) Weight balance constraint

[0098] |Wi-Wj|≤δ

[0099] Among them, δ represents the preset value of the balance constraint. During the adjustment process, the weights of different rules need to maintain a certain balance. In particular, the weight difference between key indicators should be controlled within a reasonable range to prevent a certain rule from occupying too many resources and causing scheduling imbalance.

[0100] The above system constraints ensure the reasonable distribution of weights and avoid extreme situations that may occur during the adjustment process.

[0101] like Figure 2 As shown, Figure 2 A schematic diagram of the production scheduling strategy optimization system architecture in this embodiment is shown.

[0102] The present invention also provides a production scheduling strategy optimization system based on state perception and user decision-making, comprising:

[0103] Collection module, used to collect real-time production status data;

[0104] The state evaluation module is used to standardize the collected data and convert it into a unified state vector;

[0105] A dynamic weight calculation module is used to calculate the dynamic weight adjustment amount of each dispatching rule according to the state vector;

[0106] A priority evaluation module, used to calculate the rule priority of each scheduling rule;

[0107] The weight smoothing processing module is used to smooth the dynamic weight adjustment amount to obtain the final weight adjustment value;

[0108] The execution module is used to execute the production scheduling strategy according to the rule priority and the final weight adjustment value, evaluate the actual production effect, adjust the user decision model parameters and adaptively optimize the production scheduling strategy.

[0109] The present invention provides a production scheduling strategy optimization method and system based on state perception and user decision-making. By establishing a dynamic weight adjustment mechanism based on production status, the problem of fixed weight parameters in traditional production scheduling is solved, and the adaptability and accuracy of scheduling schemes are improved. Moreover, by introducing a weight smoothing processing mechanism and a multi-dimensional effect evaluation model, the problem of unstable scheduling schemes caused by complex and changeable production environments is solved, and the continuity and stability of system adjustments are ensured. In addition, by establishing a user decision learning model, intelligent extraction and reuse of scheduling experience are realized, the defects of strong experience dependence and low efficiency of manual adjustment of scheduling parameters are overcome, and intelligent parameter adjustment suggestions are provided.

[0110] It should be pointed out that, according to the needs of implementation, the various steps described in this application can be split into more steps / components, and two or more steps or partial operations of steps can be combined into a new step to achieve the purpose of the present invention.

[0111] The order of execution of each step in the above embodiment does not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0112] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A production scheduling strategy optimization method based on state perception and user decision-making, characterized in that: The method comprises: Collect real-time production status data; Standardize the collected data and convert it into a unified state vector; Calculate the dynamic weight adjustment amount of each dispatch rule according to the state vector; Calculate the rule priority of each scheduling rule; Smoothing the dynamic weight adjustment amount to obtain the final weight adjustment value; Execute the production scheduling strategy according to the rule priority and the final weight adjustment value, evaluate the actual production effect, adjust the user decision model parameters and adaptively optimize the production scheduling strategy.

2. The production scheduling strategy optimization method based on state perception and user decision-making according to claim 1 is characterized in that: The real-time production status data includes: equipment status data, inventory status data and order status data.

3. The production scheduling strategy optimization method based on state perception and user decision-making according to claim 1 or 2, characterized in that: The state vector is specifically calculated from equipment status data, inventory status data and order status data; the equipment status data is obtained by standardized processing of equipment utilization rate, failure rate and production capacity balance; the inventory status data is obtained by standardized processing of inventory turnover rate, inventory pressure index and raw material supply risk; the order status data is obtained by standardized processing of order delay rate, delivery urgency and plan completion rate.

4. The production scheduling strategy optimization method based on state perception and user decision-making according to claim 1 is characterized in that: The dynamic weight adjustment amount is specifically obtained by multiplying the basic adjustment coefficient, the rule response function, the time series decay function, the historical cumulative influence function, and the effect evaluation function.

5. The production scheduling strategy optimization method based on state perception and user decision-making according to claim 1 is characterized in that: The rule priority is specifically calculated by combining the urgency, importance, and timeliness of each scheduling rule according to a weighted combination.

6. The production scheduling strategy optimization method based on state perception and user decision-making according to claim 1 is characterized in that: The final weight adjustment value is specifically obtained by adding the current weight adjustment value to the product of the smoothing coefficient and the dynamic weight adjustment amount.

7. The production scheduling strategy optimization method based on state perception and user decision-making according to claim 1 is characterized in that: The calculation of the dynamic weight adjustment amount and the final weight adjustment value is subject to system constraints, which include: The weight adjustment value of each dispatch rule is between 0 and 1; The sum of the weight adjustment values ​​of each dispatch rule is equal to 1; The absolute value of the dynamic weight adjustment amount of each dispatching rule is less than or equal to the preset value; The weight adjustment values ​​of each dispatching rule are subtracted from each other, and the absolute value of the difference is less than or equal to the preset value.

8. The production scheduling strategy optimization method based on state perception and user decision-making according to claim 1 is characterized in that: The user decision model is obtained in the following way: Collect user historical decision data, including production status and adjustment strategy data; Analyze user historical decision data, extract key features related to adjustment strategies, and conduct differentiated analysis of user adjustment behaviors; Establish a mapping model based on production status and user adjustment behavior; Calculate the similarity between different production status scenarios and identify historical scenarios similar to the current production status; Further improve the mapping model and establish user decision models under different production conditions; The adjusting of the user decision model parameters specifically refers to generating a recommended dynamic weight adjustment value based on the user decision model of the current production status.

9. The production scheduling strategy optimization method based on state perception and user decision-making according to claim 1 is characterized in that: The adaptive optimization of the production scheduling strategy is achieved by adjusting the dynamic weight adjustment amount of each scheduling rule and the rule priority to minimize the optimization objective function value.

10. A production scheduling strategy optimization system based on state perception and user decision-making, characterized in that: include: Collection module, used to collect real-time production status data; The state evaluation module is used to standardize the collected data and convert it into a unified state vector; A dynamic weight calculation module is used to calculate the dynamic weight adjustment amount of each dispatching rule according to the state vector; A priority evaluation module, used to calculate the rule priority of each scheduling rule; The weight smoothing processing module is used to smooth the dynamic weight adjustment amount to obtain the final weight adjustment value; The execution module is used to execute the production scheduling strategy according to the rule priority and the final weight adjustment value, evaluate the actual production effect, adjust the user decision model parameters and adaptively optimize the production scheduling strategy.

Images