Security industry scheduling planning method based on plan scheduling optimization algorithm
By improving the genetic algorithm to construct a three-dimensional coding system and dynamic penalty mechanism, the adaptability and flexibility of production scheduling in the security industry have been solved, achieving efficient and timely production management and reducing costs and resource waste.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- TIANJIN TIANDY DIGITAL TECH
- Filing Date
- 2025-11-27
- Publication Date
- 2026-04-17
AI Technical Summary
Existing planning and scheduling optimization algorithms lack adaptability and flexibility in the face of complex and ever-changing production environments and diverse product demands in the security industry, resulting in low production efficiency, serious resource waste, slow response speed, and difficulty in meeting market demands.
An improved genetic algorithm is used to construct a three-dimensional coding system, embedding security production-specific constraints. Combined with a dynamic penalty mechanism and real-time data-driven approach, it generates accurate scheduling plans, including basic scheduling tables, emergency adjustment plans, and cost optimization suggestions.
It has increased production efficiency by 20%-30%, shortened production cycle by 15%-25%, improved on-time delivery rate to over 90%, reduced raw material and equipment energy consumption costs, enhanced emergency response capabilities, and adapted to the expansion of enterprise scale and changes in product types.
Smart Images
Figure SMS_9 
Figure SMS_12 
Figure SMS_20
Abstract
Description
Technical Field
[0001] This invention belongs to the field of security planning technology, and in particular to a scheduling planning method for the security industry based on a scheduling optimization algorithm. Background Technology
[0002] In today's digital age, intelligent manufacturing has become a significant trend in global manufacturing development. As a key sector ensuring social safety and stability, the security industry is also actively integrating into this wave of intelligent manufacturing. With accelerated urbanization, increased public awareness of security, and the frequent occurrence of various security incidents, the demand in the security market is experiencing rapid growth. Traditional security manufacturing models, facing increasingly complex and volatile market demands and fierce competition, are gradually revealing numerous problems, such as low production efficiency, serious resource waste, and slow response times, making it difficult to meet the industry's development needs.
[0003] Planning and scheduling, as a core component of manufacturing, plays a decisive role in a company's production efficiency, cost control, and customer satisfaction. In the intelligent transformation, digitalization, and networked manufacturing of the security industry, optimizing planning and scheduling algorithms is of paramount importance. By employing advanced optimization algorithms, the rational allocation of production resources can be achieved, minimizing resource idleness and waste during the production process, thereby effectively reducing production costs. For example, accurate planning and scheduling ensures that raw materials, equipment, and human resources are invested in the appropriate production stages at the right time, avoiding delays and increased costs caused by resource mismatches.
[0004] Optimizing scheduling algorithms can also significantly improve production efficiency. Reasonable scheduling can reduce waiting and changeover times during production, keeping the production line running efficiently. Taking the production of security products as an example, by optimizing scheduling, rapid switching between different products can be achieved, improving equipment utilization, thereby shortening the product production cycle and responding to market demands more quickly.
[0005] Customer satisfaction is also a key factor in a company's success. Optimizing scheduling algorithms helps ensure on-time product delivery and meets customer requirements. In the security industry, customers typically have strict requirements for product delivery times, and timely delivery is crucial for ensuring the smooth progress of projects. By optimizing scheduling, companies can improve the timeliness of order delivery, enhance customer trust and satisfaction, and ultimately improve their market competitiveness.
[0006] In the field of production scheduling optimization algorithms, research abroad started earlier and has yielded fruitful results. Numerous scholars and research institutions have proposed a series of classic optimization algorithms using theories from operations research, artificial intelligence, and other disciplines. For example, genetic algorithms efficiently search the solution space to find the optimal solution by simulating natural selection and genetic mechanisms; tabu search algorithms avoid getting trapped in local optima by setting tabu lists, continuously exploring a better solution space. In practical applications, these algorithms have been widely used in industries such as automobile manufacturing and electronics manufacturing, achieving significant results. For instance, a well-known international automobile manufacturer used genetic algorithms to optimize production scheduling, resulting in a 20% increase in production efficiency and a 15% reduction in costs.
[0007] Domestic research in the field of production planning and scheduling optimization algorithms has also made significant progress in recent years. Domestic scholars have improved and innovated traditional algorithms based on the actual needs and characteristics of local enterprises, and are actively exploring new algorithms and methods. For example, some scholars have combined particle swarm optimization with simulated annealing to propose a new hybrid optimization algorithm, which has shown better performance in solving complex production planning and scheduling problems. In the area of intelligent transformation and digitalization in the security industry, domestic enterprises are also actively promoting the application and practice of related technologies. Some leading security companies have introduced intelligent manufacturing technologies to achieve digital and intelligent management of the production process, improving production efficiency and product quality.
[0008] However, current research still has some shortcomings. On the one hand, existing scheduling optimization algorithms need further improvement in adaptability and flexibility when facing the complex and ever-changing production environment and diverse product demands of the security industry. The production of security products involves various raw materials, components, and production processes, and market demand fluctuates significantly, which places higher demands on the adaptability of scheduling algorithms. Summary of the Invention
[0009] The purpose of this invention is to overcome the shortcomings of the prior art and propose a scheduling planning method for the security industry based on a planning and scheduling optimization algorithm. This method effectively solves the problems of adaptability and flexibility of production scheduling in the security industry and significantly improves production efficiency, resource utilization and system reliability.
[0010] A scheduling planning method for the security industry based on a scheduling optimization algorithm includes the following steps: Step S1: Construction of security scenario-based algorithm model and generation of schedule: Collect production data, construct a scheduling model with embedded security production-specific constraints based on an improved genetic algorithm; after training the model, output an optimized scheduling scheme. Step S2: Based on the optimized scheduling scheme output, perform corresponding security industry scheduling arrangements.
[0011] Furthermore, step S1 includes the following sub-steps: Step S1.1, Data Acquisition: Periodically collect order data, resource data, and production data through industrial sensors and the MES system; Step S1.2, Model Construction: Based on the improved genetic algorithm, a three-dimensional coding system of "order-process-resource" is designed, and equipment calibration cycle constraints, quality inspection pass rate constraints and emergency order priority constraints are embedded. Step S1.3, Model Training: Train the model using historical production data, adjust the fitness function weights using gradient descent, and introduce a dynamic penalty mechanism to penalize scheduling schemes that violate constraints. Step S1.4, Scheduling Output: Based on real-time data, output the basic scheduling table, emergency adjustment plan and cost optimization suggestions.
[0012] Furthermore, in step S1.1, the order data includes product model, customized functions, delivery date and quality standards; the resource data includes equipment capacity, calibration cycle and worker skill matrix; and the production data includes process time, quality inspection pass rate and material loss rate.
[0013] Furthermore, in step S1.2, the three-dimensional coding system is represented in the format "XYZ", where X is the order number, Y is the process number, and Z is the resource equipment number; the priority constraint for emergency orders sets a higher weight coefficient for public safety orders than for ordinary orders.
[0014] Moreover, in step S1.3, the dynamic penalty mechanism guides the algorithm's search direction by significantly reducing the fitness function value of scheduling schemes that violate constraints.
[0015] Furthermore, in step S1.4, the emergency adjustment plan includes at least a response plan for equipment failure and material shortage; the cost optimization suggestions include a raw material procurement time window and a plan for utilizing equipment energy consumption off-peak periods.
[0016] The advantages and positive effects of this invention are: I. Increased efficiency Better resource allocation: The "improved genetic algorithm - security production constraint" coupled model achieves precise matching of orders, processes and resources through a three-dimensional coding system and exclusive constraint embedding, improving equipment utilization by 20%-30% and avoiding the problem of equipment idleness and overload coexisting.
[0017] Shorter production cycle: After model training, the prediction error is controlled within 5%, and the dynamic penalty mechanism ensures scheduling compliance. The production cycle is shortened by an average of 15%-25%, which is especially suitable for the complex production process of customized security orders.
[0018] Improved on-time delivery rate: The priority constraints of emergency orders and the dynamic adjustment driven by real-time data have increased the on-time delivery rate from about 70% in the traditional model to over 90%, effectively responding to the needs of emergency orders related to public safety.
[0019] II. Enhanced Operation and Maintenance and Emergency Response Capabilities More efficient emergency response: The scheduling output includes emergency adjustment plans for 3 typical scenarios. In case of equipment failure or material shortage, there is no need to recalculate the entire schedule. The scheduling adjustment can be completed within 1-2 hours, reducing production interruption time.
[0020] More precise cost optimization: The cost optimization suggestions provide targeted raw material procurement windows and equipment energy consumption off-peak utilization plans, reducing raw material inventory costs by 15%-20% and equipment energy consumption costs by 8%-12%.
[0021] The algorithm's adaptability is continuously improved: the technology iteration method dynamically optimizes algorithm parameters based on case feedback, improving the convergence speed by 40%. It can be continuously adapted as the enterprise's production scale expands and product types increase, thus extending the technology life cycle.
[0022] III. Excellent Industry Adaptability and Promotional Value Tailored to the characteristics of security production: It is specifically embedded with security-specific constraints such as equipment calibration cycle and key process quality inspection pass rate, which is more adaptable than general scheduling algorithms and solves the scheduling pain points of customized and technology-intensive production of security products.
[0023] Controllable implementation costs: Based on data collected from existing industrial sensors and MES systems, no large-scale hardware modifications are required, allowing small and medium-sized security companies to deploy at low cost.
[0024] Smoother supply chain collaboration: Suppliers only obtain material demand data, achieving precise matching of supply and demand while ensuring data security, reducing downtime due to material shortages or inventory backlog. Detailed Implementation
[0025] The technical problem solved by this invention is achieved through the following technical solution: The first method used was a comparative experiment. In the same production scenario, traditional scheduling methods and optimization algorithms were run separately, and the differences were quantified through eight indicators, including production cycle, equipment utilization rate, and on-time delivery rate. The second method was the technology iteration method, which optimized the algorithm parameters based on case feedback data. For example, the crossover probability and mutation probability of the genetic algorithm were adjusted, which improved the convergence speed by 40%.
[0026] Research innovations: A coupled model of "improved genetic algorithm - security production constraints" is proposed, which realizes customized scheduling in four steps: Step 1: Data Acquisition. Collect three types of core data through industrial sensors and the MES system: order data (product model, customized functions, delivery date, quality standards), resource data (equipment capacity, calibration cycle, worker skill matrix), and production data (process time, quality inspection pass rate, material loss rate). The data sampling frequency is set to once every 5 minutes to ensure real-time performance.
[0027] Step 2: Model construction. Based on the improved genetic algorithm, a three-dimensional coding system of "order-process-resource" is designed (e.g., the code "05-03-08" represents "the 3rd process (lens calibration) of the 5th customized security monitoring order is assigned to the 8th calibration device"). At the same time, security production-specific constraints are embedded: equipment calibration cycle constraints (e.g., shutdown for calibration every 200 hours), quality inspection qualification constraints (the qualification rate of key processes must reach 99.5% or more), and emergency order priority constraints (the priority weight of public safety orders is set to 0.6).
[0028] Step 3: Model training. The model is trained using the company's historical production data for the past 12 months (including 500+ orders and 3000+ process records). The fitness function weights are adjusted using gradient descent to keep the model prediction error within 5%. A "dynamic penalty mechanism" is introduced during training to apply penalty coefficients to scheduling schemes that violate security constraints, thereby improving the model's compliance.
[0029] Step 4: Scheduling output. The model generates three types of scheduling results based on real-time data: basic scheduling table (including process allocation, equipment usage time, and personnel configuration), emergency adjustment plan (for three typical scenarios of equipment failure and material shortage), and cost optimization suggestions (such as raw material procurement time window and equipment energy consumption off-peak period utilization plan).
[0030] The specific formula and implementation process of the improved genetic algorithm: 1. Encoding Structure Chromosome representation: One chromosome represents a complete production scheduling plan.
[0031]
[0032] Each gene Represents a process allocation Gene analysis: Order number; This order The first in One process; : The resource number assigned to this process (the resource pool contains production equipment and operators, which can be associated through a mapping table).
[0033] Chromosome length: This refers to the total number of steps in all orders. Each step in an order must appear exactly once in the chromosome.
[0034] 2. Fitness Function Taking into account production cycle, cost, resource utilization, and constraint satisfaction, the following definition is provided:
[0035] fitness value The larger the value, the better the scheduling scheme C is.
[0036] The total completion time of the schedule, i.e., the end time of the last process in all orders;
[0037] The historical maximum completion time used for normalization; Dynamic update mechanism: That is, take the larger of the historical maximum value and the current maximum completion time of the population.
[0038] Average resource utilization rate.
[0039] Average equipment utilization rate; in It is a resource Total working time within the scheduling cycle.
[0040] : Normalized total cost, in the range [0, 1].
[0041]
[0042]
[0043] Energy consumption cost based on electricity price during equipment operation.
[0044] : Staff hour cost.
[0045] Material costs (may include premiums for emergency purchases).
[0046] The lower and upper bounds of the cost, estimated based on historical data and simple heuristics, are used for normalization.
[0047] Dynamic penalty items (see below); Online adjustment of weighting coefficients:
[0048]
[0049]
[0050] Then normalization is performed:
[0051] This is the learning rate.
[0052] This represents the average value of the current population. The target utilization rate is (e.g., 85%).
[0053] 3. Dynamic Penalty Mechanism Punish individuals who violate security-specific restrictions:
[0054] in:
[0055] : equipment Cumulative runtime since the last calibration (calculated from the schedule).
[0056] Maximum permissible runtime (e.g., 200 hours). This is a hard constraint. Penalties for quality risks in key processes
[0057] The set of all processes marked as "critical". : Indicator function, returns 1 if the condition is true, otherwise returns 0.
[0058] The historical pass rate of the equipment assigned to this process.
[0059] Quality pass rate threshold (e.g., 0.995). This is a soft constraint that encourages the use of high-quality equipment.
[0060] Penalty for delays in urgent orders.
[0061] : : Collection of urgent orders.
[0062] Orders in the scheduling plan Delivery date.
[0063] :Order The required delivery date. This is a hard constraint.
[0064] Severity of punishment: These coefficients should be much larger than the typical values of other terms in the fitness function to ensure that individuals with serious violations are decisively eliminated.
[0065] 4. Genetic operations and parameter adaptation Select: Tournament Selection (tournament size=5); Crossover: POX (Precedence Preserving Order-based Crossover) Mutation: Randomly swapping the resource allocation of two non-conflicting processes, or reallocating legal resources to a process; Parameter adaptive adjustment:
[0066]
[0067] Standard deviation of fitness in contemporary populations.
[0068] The maximum and minimum fitness values of the current population. Overall Algorithm Flow 1. Initialization Read real-time data such as orders, process routes, and resource status (equipment availability, personnel shifts); Setting genetic algorithm parameters (initial) , Population size Maximum Algebra ); Generate the initial population: Randomly generate valid chromosomes, ensuring that all processes are assigned and satisfy the basic sequence; Iterative optimization: a. Decoding and Scheduling Generation: Decode each chromosome C into a specific scheduling timetable. This is usually done by a scheduling generator, which calculates the start and end times (T(C)) of each process based on the process-resource matching order in the chromosome, combined with process duration, resource calendar, and other information.
[0069] b. Fitness assessment: Calculate F(C) for each individual according to the formula in step 2.
[0070] c. Elite Preservation: The best individuals of the present generation are directly preserved to the next generation to prevent the loss of excellent genes.
[0071] d. Selection, crossover, and mutation: Generate the offspring population based on the operators in step 4.
[0072] e. Parameter adaptation: based on the current population Adjustment and .
[0073] f. Weight Adaptive: Fine-tune ω1, ω2, ω3 online based on the average performance of the current population.
[0074] 3. Termination and Output Termination condition: Reaching the maximum algebra Or, the optimal solution does not show significant improvement over multiple generations.
[0075] Output: Decode the optimal chromosome in the final generation to generate a basic scheduling table (Gantt chart, process list). Derivative output: Emergency adjustment plan: Simulate equipment failure (remove the equipment and reschedule several processes) and material shortage (postpone affected orders), and quickly apply optimization algorithms to generate 2-3 feasible adjustment plans.
[0076] Cost optimization recommendations: Analyze the optimal scheduling scheme and identify cost components. For example, identify whether high-energy-consuming processes are concentrated during peak electricity price periods and suggest shifting them to off-peak periods; analyze material inventory time and suggest purchasing before specific time windows to take advantage of discounts.
[0077] It should be emphasized that the embodiments described in this invention are illustrative rather than limiting. Therefore, this invention includes, but is not limited to, the embodiments described in the specific implementation. Any other implementations derived by those skilled in the art based on the technical solutions of this invention are also within the scope of protection of this invention.
Claims
1. A scheduling planning method for the security industry based on a planning and scheduling optimization algorithm, characterized in that: Includes the following steps: Step S1: Construction of security scenario-based algorithm model and generation of schedule: Collect production data, construct a scheduling model with embedded security production-specific constraints based on an improved genetic algorithm; after training the model, output an optimized scheduling scheme. Step S2: Based on the optimized scheduling scheme output, perform corresponding security industry scheduling arrangements.
2. The method according to claim 1, characterized in that, Step S1 includes the following sub-steps: Step S1.1, Data Acquisition: Periodically collect order data, resource data, and production data through industrial sensors and the MES system; Step S1.2, Model Construction: Based on the improved genetic algorithm, a three-dimensional coding system of "order-process-resource" is designed, and equipment calibration cycle constraints, quality inspection pass rate constraints and emergency order priority constraints are embedded. Step S1.3, Model Training: Train the model using historical production data, adjust the fitness function weights using gradient descent, and introduce a dynamic penalty mechanism to penalize scheduling schemes that violate constraints. Step S1.4, Scheduling Output: Based on real-time data, output the basic scheduling table, emergency adjustment plan and cost optimization suggestions.
3. The method according to claim 2, characterized in that: In step S1.1, the order data includes product model, customized functions, delivery date and quality standards; the resource data includes equipment capacity, calibration cycle and worker skill matrix; and the production data includes process time, quality inspection pass rate and material loss rate.
4. The method according to claim 2, characterized in that: In step S1.2, the three-dimensional coding system is represented in the format "XYZ", where X is the order number, Y is the process number, and Z is the resource equipment number; the priority constraint of emergency orders sets a higher weight coefficient for public safety orders than for ordinary orders.
5. The method according to claim 2, characterized in that: In step S1.3, the dynamic penalty mechanism guides the algorithm's search direction by significantly reducing the fitness function value of scheduling schemes that violate constraints.
6. The method according to claim 2, characterized in that: In step S1.4, the emergency adjustment plan includes at least a response plan for equipment failure and material shortage; the cost optimization suggestions include a raw material procurement time window and a plan for utilizing equipment energy consumption off-peak periods.