A method for solving complex product assembly sequence planning problem considering human factors

Abstract

The application relates to a method for solving a complex product assembly sequence planning problem considering human factors, wherein the method comprises: comprehensively introducing cognitive, physiological, psychological, skill and organizational human factors in the complex product assembly sequence planning process, and optimizing and solving the assembly sequence based on genetic programming. By effectively deciding the human factor data acquisition mode, the state data of workers are acquired, the assembly sequence is adjusted in combination with the constraint of the human factor, and finally the optimal assembly sequence planning considering the state of workers is realized, so that the efficiency and reliability of the assembly process are improved. Therefore, the problems that the prior art relies on the experience of engineers for manual planning, human factors are difficult to be comprehensively considered, and modern complex product manufacturing requirements cannot be met are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of complex product assembly, and particularly relates to a method for solving complex product assembly sequence planning problem considering human factors. BACKGROUND

[0002] A complex product refers to a product with complex customer demand, complex system composition, complex product technology, complex manufacturing process, complex test and maintenance, complex project management and complex working environment, such as a spacecraft, an airplane, a ship, a complex electromechanical product and the like. The assembly process of a complex product has a high requirement for assembly sequence planning due to a large number of component types and complex structure.

[0003] Traditional assembly sequence planning usually relies on the experience of engineers for manual planning, which is time-consuming and easily affected by subjective factors, and it is difficult to achieve efficient assembly. In addition, the state changes of workers in the assembly process (such as fatigue, skill difference, etc.) have a significant impact on assembly efficiency and quality, but the traditional method often ignores these human factors, resulting in instability and potential safety hazards in the assembly process. With the increasing requirements of modern industrial market on the assembly precision and efficiency of complex products, and the increasing complexity of assembly, the traditional manual planning method is difficult to meet the high standard requirements of modern complex product manufacturing.

[0004] In summary, the prior art relies on the experience of engineers for manual planning, and it is difficult to fully consider human factors, and it cannot meet the requirements of modern complex product manufacturing, which needs to be solved urgently. SUMMARY

[0005] The present application provides a method for solving complex product assembly sequence planning problem considering human factors, to solve the problems that the prior art relies on the experience of engineers for manual planning, it is difficult to fully consider human factors, and it cannot meet the requirements of modern complex product manufacturing.

[0006] The first aspect embodiment of the application provides a method for solving a complex product assembly sequence planning problem considering human factors, comprising the following steps: based on a preset human factor identification framework, a human factor library of a target complex product assembly field is established, and a priority evaluation model is constructed according to a preset analytic hierarchy process strategy, and the priority evaluation model is used to calculate the comprehensive score of each human factor in the human factor library, wherein the human factor library includes cognitive factors, physiological factors, psychological factors, skill factors and organizational factors; based on a pre-constructed data acquisition method library, human factor data corresponding to each human factor is acquired, and mathematical modeling is performed on each human factor using the human factor data to obtain a complex product assembly sequence planning model corresponding to the target complex product assembly field; the start-up face condition of the complex product assembly sequence planning model is judged, if the start-up face condition is a single start-up face condition, then based on a preset heuristic algorithm, the complex product assembly sequence planning model under the single start-up face condition is solved to obtain a corresponding target assembly sequence; if the start-up face condition is a multi-start-up face condition, then the assembly time, assembly cost, each human factor and multiple constraint conditions of the complex product assembly sequence planning model under the multi-start-up face condition are determined, and a fitness function is constructed according to the assembly time, the assembly cost, each human factor and the multiple constraint conditions, so as to solve the complex product assembly sequence planning model under the multi-start-up face condition by using the comprehensive score, the fitness function and a preset hybrid genetic programming algorithm, and obtain a corresponding target assembly sequence.

[0007] Optionally, in an embodiment of the application, the human factor library of the target complex product assembly field is established based on the preset human factor identification framework, and the priority evaluation model is constructed according to the preset analytic hierarchy process strategy, and the comprehensive score of each human factor in the human factor library is calculated by using the priority evaluation model, comprising: based on the actual work requirements of the target complex product assembly field and a plurality of preset factor identification theories, the human factor identification framework is constructed to establish the human factor library of the target complex product assembly field according to the human factor identification framework, wherein the plurality of factor identification theories include human factors engineering theory, work design theory, SHEL model, operation risk management theory and organizational behavior theory; the relative importance score of each pair of human factors in the human factor library is calculated to construct a judgment matrix, and the maximum eigenvalue and consistency index of the judgment matrix are calculated; based on the maximum eigenvalue and the consistency index, the consistency ratio is calculated, and the consistency of each pair of human factors is tested according to the consistency ratio to obtain a corresponding consistency test result; the data of each column of the judgment matrix is normalized to calculate the weight of each human factor, and based on the weight and the consistency test result, the comprehensive score of each human factor in the human factor library is determined.

[0008] Optionally, in an embodiment of the present application, the human factor data corresponding to each human factor is collected based on a pre-constructed data collection method library, comprising: determining a cognitive factor data collection method, a physiological factor data collection method, a psychological factor data collection method, a skill factor data collection method and an organizational factor data collection method corresponding to the cognitive factor, the physiological factor, the psychological factor, the skill factor and the organizational factor respectively; and constructing the data collection method library according to the cognitive factor data collection method, the physiological factor data collection method, the psychological factor data collection method, the skill factor data collection method and the organizational factor data collection method.

[0009] Optionally, in an embodiment of the present application, the human factor data corresponding to each human factor is collected based on a pre-constructed data collection method library, comprising: determining a cognitive factor data collection method, a physiological factor data collection method, a psychological factor data collection method, a skill factor data collection method and an organizational factor data collection method corresponding to the cognitive factor, the physiological factor, the psychological factor, the skill factor and the organizational factor respectively; and constructing the data collection method library according to the cognitive factor data collection method, the physiological factor data collection method, the psychological factor data collection method, the skill factor data collection method and the organizational factor data collection method.

[0010] Optionally, in one embodiment of the present application, the determination of the assembly time, the assembly cost, each human factor and multiple constraint conditions of the complex product assembly sequence planning model under the multi-operation face condition, and the construction of the fitness function according to the assembly time, the assembly cost, each human factor and the multiple constraint conditions, so as to solve the complex product assembly sequence planning model under the multi-operation face condition by the comprehensive score, the fitness function and a preset hybrid genetic programming algorithm to obtain a corresponding target assembly sequence, comprises: reading and analyzing multiple constraint conditions of parallel assembly sequence planning, wherein the multiple constraint conditions include assembly process constraints, resource constraints, space constraints, maximum operation face number and human factor constraints; determining a factor score and a weight coefficient of each important human factor corresponding to each complex product assembly task, and constructing a human factor evaluation function by the factor score and the weight coefficient, so as to construct the fitness function according to the assembly time, the assembly cost, the human factor evaluation function and the multiple constraint conditions, and determine an expression form of the assembly sequence planning result by the fitness function, and perform a neighborhood search operation on the assembly sequence planning result to obtain a corresponding initial population; based on the hybrid genetic programming algorithm, screening multiple individuals meeting a preset fitness requirement from the initial population, and performing a crossover operation on each individual in the multiple individuals and a parent individual in the initial population corresponding to the individual according to a preset crossover strategy to obtain multiple new child individuals; based on a preset mutation probability, performing a mutation operation on the multiple new child individuals, and performing fitness evaluation on the multiple new child individuals after the mutation operation to obtain a fitness function value corresponding to each new child individual in the multiple new child individuals, and iteratively performing the screening, the crossover, the mutation and the fitness evaluation operations until a preset termination condition is met, so as to obtain a target assembly sequence meeting a preset fitness value requirement.

[0011] Optionally, in one embodiment of the present application, after solving the complex product assembly sequence planning model under the multi-operation face condition by the comprehensive score, the fitness function and a preset hybrid genetic programming algorithm to obtain a corresponding target assembly sequence, the method further comprises: based on the data acquisition mode library, acquiring current human factor data of workers in the process of performing the complex product assembly task; dynamically adjusting a human factor evaluation function in the fitness function according to the current human factor data, so as to solve the complex product assembly sequence planning model under the multi-operation face condition by using the dynamically adjusted human factor evaluation function.

[0012] Optionally, in one embodiment of the present application, the mathematical expression of the fitness function is:

[0013] Fitness(X)=θ·T total(X) + β · C total (X) + γ · F human (X) + λ · Penalty(X)

[0014] wherein, X represents an assembly sequence scheme; T total (X) represents an assembly time of the assembly sequence scheme X; C total (X) represents an assembly cost; F human (X) represents a human factor evaluation function; Penalty(X) represents a penalty term for violating a constraint condition; θ represents a weight coefficient corresponding to the assembly time; β represents a weight coefficient corresponding to the assembly cost; γ represents a weight coefficient corresponding to the human factor evaluation function; and λ represents a weight coefficient corresponding to the penalty term.

[0015] The second aspect embodiment of the present application provides a device for solving a complex product assembly sequence planning problem considering human factors, comprising: a modeling module, configured to establish a human factor library in a target complex product assembly field based on a preset human factor identification framework, and construct a priority evaluation model according to a preset analytic hierarchy process strategy, and calculate a comprehensive score of each human factor in the human factor library by using the priority evaluation model, wherein the human factor library comprises cognitive factors, physiological factors, psychological factors, skill factors and organizational factors; a collection module, configured to collect human factor data corresponding to each human factor based on a pre-constructed data collection mode library, and perform mathematical modeling on each human factor by using the human factor data, to obtain a complex product assembly sequence planning model corresponding to the target complex product assembly field; a first solving module, configured to judge a start face condition of the complex product assembly sequence planning model, and if the start face condition is a single start face condition, solve the complex product assembly sequence planning model under the single start face condition based on a preset heuristic algorithm, to obtain a corresponding target assembly sequence; and a second solving module, configured to if the start face condition is a multi-start face condition, determine an assembly time, an assembly cost, each human factor and a plurality of constraint conditions of the complex product assembly sequence planning model under the multi-start face condition, and construct a fitness function according to the assembly time, the assembly cost, each human factor and the plurality of constraint conditions, to solve the complex product assembly sequence planning model under the multi-start face condition by using the comprehensive score, the fitness function and a preset hybrid genetic programming algorithm, to obtain a corresponding target assembly sequence.

[0016] Optionally, in an embodiment of the present application, the modeling module comprises: a first constructing unit configured to construct the human factor identification framework based on actual work requirements of the target complex product assembly field and preset factor identification theories, and to establish the human factor library of the target complex product assembly field according to the human factor identification framework, wherein the factor identification theories comprise human factor engineering theory, work design theory, SHEL model, operation risk management theory and organizational behavior theory; a scoring unit configured to score relative importance of each pair of human factors in the human factor library to construct a judgment matrix, and to calculate a maximum eigenvalue and a consistency index of the judgment matrix; a consistency checking unit configured to calculate a consistency ratio based on the maximum eigenvalue and the consistency index, and to perform consistency checking on the each pair of human factors according to the consistency ratio to obtain a corresponding consistency checking result; and a normalization unit configured to normalize each column of data of the judgment matrix to calculate a weight of each human factor, and to determine a comprehensive score of each human factor in the human factor library based on the weight and the consistency checking result.

[0017] Optionally, in an embodiment of the present application, the collecting module comprises: a determining unit configured to determine a cognitive factor data collection mode, a physiological factor data collection mode, a psychological factor data collection mode, a skill factor data collection mode and an organizational factor data collection mode corresponding to the cognitive factor, the physiological factor, the psychological factor, the skill factor and the organizational factor respectively; and a second constructing unit configured to construct the data collection mode library according to the cognitive factor data collection mode, the physiological factor data collection mode, the psychological factor data collection mode, the skill factor data collection mode and the organizational factor data collection mode.

[0018] Optionally, in an embodiment of the present application, the collecting module further comprises: a quantifying unit configured to quantify attention, memory and decision-making ability of a worker in a complex product assembly task execution process of the target complex product assembly field, so as to construct the cognitive factor mathematical model according to the quantified attention, memory and decision-making ability of the worker; an estimating unit configured to acquire working time, rest time and at least one physiological index of the worker in the complex product assembly task execution process, determine fatigue degree of the worker according to the working time and the rest time, and estimate physical state of the worker through the at least one physiological index, so as to construct the physiological factor mathematical model according to the fatigue degree and the physical state; a first acquiring unit configured to quantify emotional state of the worker in the complex product assembly task execution process through a preset emotional score scale, acquire at least one physiological parameter of the worker, and quantify stress level of the worker by using the at least one physiological parameter and a preset subjective report, so as to construct the psychological factor mathematical model based on the quantified emotional state and stress level; a calculating unit configured to acquire average time and reference time of the worker in completing the same complex product assembly task, and calculate skill level of the worker according to the average time and the reference time; a second acquiring unit configured to acquire participating frequency of the worker in assembly tasks that satisfy a preset similarity requirement with the complex product assembly task, and construct the skill factor mathematical model according to the skill level and the participating frequency; and a third constructing unit configured to determine training degree, rule compliance degree and team cooperation efficiency of the worker, and construct the organization factor mathematical model according to the training degree, the rule compliance degree and the team cooperation efficiency.

[0019] Optionally, in an embodiment of the present application, the second solving module comprises: a reading unit configured to read and parse a plurality of constraint conditions of the parallel assembly sequence planning, wherein the plurality of constraint conditions comprise assembly process constraints, resource constraints, space constraints, maximum open face quantity, and human factor constraints; a neighborhood searching unit configured to determine a factor score and a weight coefficient of each important human factor corresponding to each complex product assembly task, and construct a human factor evaluation function by using the factor score and the weight coefficient, so as to construct the fitness function according to the assembly time, the assembly cost, the human factor evaluation function, and the plurality of constraint conditions, and determine an expression form of an assembly sequence planning result by using the fitness function, and perform a neighborhood search operation on the assembly sequence planning result to obtain a corresponding initial population; an algorithm processing unit configured to filter a plurality of individuals satisfying a preset fitness requirement from the initial population based on the hybrid genetic programming algorithm, and perform a crossover operation on each individual in the plurality of individuals and a parent individual in the initial population corresponding to the individual according to a preset crossover strategy to obtain a plurality of new child individuals; and an iteration unit configured to perform a mutation operation on the plurality of new child individuals based on a preset mutation probability, and perform fitness evaluation on the plurality of new child individuals after the mutation operation to obtain a fitness function value corresponding to each new child individual in the plurality of new child individuals, and iteratively perform the filtering, the crossover, the mutation, and the fitness evaluation operations until a preset termination condition is satisfied, so as to obtain a target assembly sequence satisfying a preset fitness value requirement.

[0020] Optionally, in an embodiment of the present application, the method further comprises: after the complex product assembly sequence planning model in the multi-open face condition is solved by using the comprehensive score, the fitness function, and the preset hybrid genetic programming algorithm to obtain a corresponding target assembly sequence, acquiring current human factor data of workers in a process of performing the complex product assembly task based on the data acquisition mode library; and dynamically adjusting a human factor evaluation function in the fitness function according to the current human factor data, so as to solve the complex product assembly sequence planning model in the multi-open face condition by using the dynamically adjusted human factor evaluation function.

[0021] Optionally, in an embodiment of the present application, the mathematical expression of the fitness function is:

[0022] Fitness(X)=θ·T total (X)+β·C total (X)+γ·F human (X)+λ·Penalty(X)

[0023] wherein X represents an assembly sequence scheme; T total(X) represents the assembly time of the assembly sequence scheme X; C total (X) represents the assembly cost; F human (X) represents the human factor evaluation function; Penalty(X) represents the penalty term for violating the constraint condition; θ represents the weight coefficient corresponding to the assembly time; β represents the weight coefficient corresponding to the assembly cost; γ represents the weight coefficient corresponding to the human factor evaluation function; and λ represents the weight coefficient corresponding to the penalty term.

[0024] The third aspect of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for solving the complex product assembly sequence planning problem considering human factors as described in the above embodiments.

[0025] The fourth aspect of the present application provides a computer readable storage medium, which stores a computer program executable by a processor to implement the method for solving the complex product assembly sequence planning problem considering human factors as described above.

[0026] Therefore, the embodiments of the present application have the following beneficial effects:

[0027] The embodiment of the present application can establish a human factor library in the target complex product assembly field based on a preset human factor identification framework, construct a priority evaluation model according to a preset analytic hierarchy process strategy, and calculate the comprehensive score of each human factor in the human factor library by using the priority evaluation model, wherein the human factor library includes cognitive factors, physiological factors, psychological factors, skill factors and organizational factors; based on a pre-constructed data acquisition mode library, human factor data corresponding to each human factor is acquired, and each human factor is mathematically modeled by using the human factor data to obtain a complex product assembly sequence planning model corresponding to the target complex product assembly field; based on a preset heuristic algorithm, the complex product assembly sequence planning model under the condition of a single work face is solved to obtain a corresponding target assembly sequence; the assembly time, assembly cost, each human factor and multiple constraint conditions of the complex product assembly sequence planning model under the condition of multiple work faces are determined, and a fitness function is constructed according to the assembly time, assembly cost, each human factor and multiple constraint conditions to solve the complex product assembly sequence planning model under the condition of multiple work faces by using the comprehensive score, the fitness function and a preset hybrid genetic programming algorithm to obtain a corresponding target assembly sequence. The assembly sequence planning method of the present application can efficiently and comprehensively consider human factors, and can realize the planning of the optimal assembly sequence under complex constraint conditions, thereby improving the assembly efficiency, reducing the cost, and ensuring the assembly quality and worker safety. Thus, the problems that the prior art relies on the experience of engineers for manual planning, cannot comprehensively consider human factors, and cannot meet the requirements of modern complex product manufacturing are solved.

[0028] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0029] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, including the accompanying drawings, wherein:

[0030] Figure 1 A flowchart of a complex product assembly sequence planning problem solving method considering human factors according to an embodiment of the present application;

[0031] Figure 2 An example output schematic diagram of a problem solving example considering the maximum work face of one worker according to an embodiment of the present application;

[0032] Figure 3 An example output schematic diagram of a problem solving example considering the maximum work face of five assembly workers according to an embodiment of the present application;

[0033] Figure 4Fig. 1 is an example diagram of a device for solving a complex product assembly sequence planning problem considering human factors according to an embodiment of the present application;

[0034] Figure 5 Fig. 1 is an example diagram of a device for solving a complex product assembly sequence planning problem considering human factors according to an embodiment of the present application;

[0035] In the drawings: 10 - device for solving a complex product assembly sequence planning problem considering human factors; 100 - modeling module, 200 - acquisition module, 300 - first solving module, 400 - second solving module; 501 - memory, 502 - processor, 503 - communication interface. DETAILED DESCRIPTION

[0036] Embodiments of the present application are described in detail below with reference to the attached drawing figures, wherein the same or like reference numerals are used throughout the drawing figures to refer to the same or like elements or elements having the same or similar functionality. The embodiments described below are exemplary and are intended to be illustrative of the present application and are not to be understood as limiting of the present application.

[0037] A human factor consideration complex product assembly sequence planning problem solving method of an embodiment of the present application is described below with reference to the accompanying drawings. In view of the problems mentioned in the background art above, the present application provides a human factor consideration complex product assembly sequence planning problem solving method. In the method, a human factor library for a target complex product assembly field is established based on a preset human factor identification framework, a priority evaluation model is constructed according to a preset analytic hierarchy process strategy, and the comprehensive score of each human factor in the human factor library is calculated using the priority evaluation model, wherein the human factor library includes cognitive factors, physiological factors, psychological factors, skill factors, and organizational factors. Based on a pre-constructed data acquisition mode library, human factor data corresponding to each human factor is collected, and each human factor is mathematically modeled using the human factor data to obtain a complex product assembly sequence planning model corresponding to the target complex product assembly field. Based on a preset heuristic algorithm, the complex product assembly sequence planning model under a single work surface condition is solved to obtain a corresponding target assembly sequence. The assembly time, assembly cost, each human factor, and multiple constraint conditions of the complex product assembly sequence planning model under a multi-work surface condition are determined, and a fitness function is constructed according to the assembly time, assembly cost, each human factor, and multiple constraint conditions to solve the complex product assembly sequence planning model under the multi-work surface condition by using the comprehensive score, fitness function, and preset hybrid genetic programming algorithm to obtain a corresponding target assembly sequence. The present application can efficiently and comprehensively consider human factors in assembly sequence planning, and can realize optimal assembly sequence planning under complex constraint conditions, thereby improving assembly efficiency, reducing cost, and ensuring assembly quality and worker safety. Thus, the problems of relying on the experience of engineers for manual planning, being difficult to comprehensively consider human factors, and being unable to meet the requirements of modern complex product manufacturing in the prior art are solved.

[0038] Specifically, Figure 1 A flowchart of a human factor consideration complex product assembly sequence planning problem solving method provided by an embodiment of the present application.

[0039] As Figure 1 shown, the human factor consideration complex product assembly sequence planning problem solving method includes the following steps:

[0040] In step S101, a human factor library for a target complex product assembly field is established based on a preset human factor identification framework, a priority evaluation model is constructed according to a preset analytic hierarchy process strategy, and the comprehensive score of each human factor in the human factor library is calculated using the priority evaluation model to obtain multiple important human factors in the human factor library through the comprehensive score, wherein the human factor library includes cognitive factors, physiological factors, psychological factors, skill factors, and organizational factors.

[0041] Different from the traditional assembly sequence planning method, in order to scientifically and systematically identify all human factors that need to be considered in the complex product assembly field, the embodiment of the present application can systematically introduce the cognitive, physiological, psychological, skill and organizational factors into the priority evaluation model, so that the assembly sequence is more in line with the actual ability and state of the workers, and the above-mentioned human factors are ensured to be comprehensively identified and analyzed.

[0042] Optionally, in an embodiment of the present application, based on the preset human factor identification framework, a human factor library of the target complex product assembly field is established, and a priority evaluation model is constructed according to the preset analytic hierarchy process strategy, and the priority evaluation model is used to calculate the comprehensive score of each human factor in the human factor library, including: based on the actual work requirements of the target complex product assembly field and the preset multiple factor identification theories, a human factor identification framework is constructed, so as to establish a human factor library of the target complex product assembly field according to the human factor identification framework, wherein the multiple factor identification theories include human factors engineering theory, job design theory, SHEL model, operation risk management theory and organizational behavior theory; the relative importance score of each pair of human factors in the human factor library is calculated to construct a judgment matrix, and the maximum eigenvalue and consistency index of the judgment matrix are calculated; based on the maximum eigenvalue and the consistency index, the consistency ratio is calculated, and the consistency of each pair of human factors is tested according to the consistency ratio to obtain the corresponding consistency test result; the data of each column of the judgment matrix is normalized to calculate the weight of each human factor, and based on the weight and the consistency test result, the comprehensive score of each human factor in the human factor library is determined.

[0043] It should be noted that in the process of identifying human factors, the embodiment of the present application first needs to establish a comprehensive framework to guide the systematic identification of factors. The embodiment of the present application can use human factors engineering (HFE), job design theory (JCM), SHEL model (software, hardware, environment, liveware), operation risk management theory, and organizational behavior theory as the basic theory of factor identification, and combine the actual work of complex product assembly to build a complete framework from cognitive factors, physiological factors, psychological factors, skill factors, and organizational factors, and evaluate the influence of human factors on complex product assembly from multiple angles, wherein various human factors are as follows:

[0044] 1. Cognitive factors: such as cognitive load of work tasks, understanding of assembly process and decision-making ability;

[0045] 2. Physiological factors: such as fatigue state, work load, physiological comfort, etc.

[0046] 3. Psychological factors: including stress, motivation, satisfaction and mental health;

[0047] 4. Skill factors: the matching degree between worker's skill and assembly task complexity;

[0048] 5. Organizational factors: such as organization's rules and regulations, skill training, management style, etc.

[0049] This overall framework can be iterated and tailored according to the actual situation of different organizations, and the overall goal is to be able to systematically and comprehensively establish the human factor library that needs to be considered in this field.

[0050] After that, the embodiments of the present application can classify all the identified human factors, and evaluate the priority according to the influence degree on the assembly process; the embodiments of the present application can establish an evaluation model through the analytic hierarchy process (AHP) to score and sort different factors, determine the key human factors that need to be focused on and optimized according to the resource constraints and specific goals of the specific assembly task, and the specific process is as follows:

[0051] 1. Constructing the judgment matrix:

[0052] For each pair of factors, the relative importance score is obtained by using the scale method to construct the judgment matrix (A), as shown in the following formula:

[0053]

[0054] For example, the importance score (a ij ) of factors (i) and (j) is (a ij = 3), which means that factor (i) is more important than factor (j), then

[0055] 2. Consistency check:

[0056] Calculate the maximum eigenvalue (λ max ) and the consistency index (CI) of the judgment matrix:

[0057]

[0058] Where (n) is the number of factors, then calculate the consistency ratio (CR):

[0059]

[0060] Where (RI) is the random consistency index; if (CR < 0.1), the consistency is acceptable;

[0061] 3. Calculate the weight:

[0062] The weight of each factor is calculated by normalizing each column of the judgment matrix; the weight (w i ) can be obtained by normalizing the sum of elements in each row;

[0063] 4. Comprehensive score:

[0064] The weights of each level are integrated to obtain the comprehensive score of each factor on the target.

[0065] Therefore, the embodiments of the present application can provide a set of steps and tools for identifying and evaluating human factors in the assembly process of complex products, thereby ensuring that the human factor library can consider the most important factors when resources are limited or the assembly task is targeted.

[0066] In step S102, based on the pre-constructed data collection method library, a plurality of important human factor data corresponding to each important human factor in a plurality of important human factors is collected, and a plurality of important human factor data is used to mathematically model each important human factor to obtain a complex product assembly sequence planning model corresponding to the target complex product assembly field.

[0067] After the human factor identification is completed, the embodiments of the present application further need to enrich and improve the list of human factors through data collection and investigation. Since each dimension of human factors has its appropriate collection method, the embodiments of the present application need to construct a large category of characteristics for each type of human factor to establish a data collection method library.

[0068] Optionally, in an embodiment of the present application, based on the pre-constructed data collection method library, a plurality of important human factor data corresponding to each important human factor in a plurality of important human factors is collected, including: respectively determining a cognitive factor data collection method, a physiological factor data collection method, a psychological factor data collection method, a skill factor data collection method, and an organizational factor data collection method corresponding to a cognitive factor, a physiological factor, a psychological factor, a skill factor, and an organizational factor; and constructing a data collection method library according to the cognitive factor data collection method, the physiological factor data collection method, the psychological factor data collection method, the skill factor data collection method, and the organizational factor data collection method.

[0069] In the embodiments of the present application, the data collection methods corresponding to the cognitive factor, the physiological factor, the psychological factor, the skill factor, and the organizational factor can be determined respectively, as follows:

[0070] 1. Data collection of cognitive factors:

[0071] The cognitive factor mainly involves the understanding, decision-making, and cognitive load of workers when performing assembly tasks. The embodiments of the present application can collect cognitive factor data by the following methods:

[0072] (1) Questionnaires and cognitive load test tools: Use standardized questionnaires such as NASA-TLX (Task Load Index) to assess workers' subjective cognitive load when performing tasks;

[0073] (2) Cognitive Task Analysis (CTA): Through detailed interviews and observations of workers, understand their thinking process and cognitive strategies when performing tasks;

[0074] (3) Laboratory simulation tests: Measure workers' information processing speed and working memory capacity by simulating assembly environments. Test tasks may include reaction time tasks and memory tasks to assess workers' cognitive abilities;

[0075] (4) Eye tracking: Use eye tracking devices to collect data on workers' fixation points and fixation times during assembly tasks to determine information processing load and attention allocation.

[0076] 2. Data collection of physiological factors:

[0077] Physiological factors include workers' fatigue state, work load, and physiological comfort, etc. Data can be collected through the following ways:

[0078] (1) Wearable physiological sensors: Use wearable devices such as heart rate monitors, skin conductance sensors, wristbands, etc. to monitor workers' heart rate variability (HRV), skin electrical activity, etc. Physiological indicators are used to assess fatigue state and stress level;

[0079] (2) Work environment monitoring equipment: Use thermometers, noise measuring instruments, illuminometers, etc. to collect physical parameters of the work environment to assess the impact of the environment on workers' physiological comfort;

[0080] (3) Motion tracking system: Through motion capture technology, record workers' posture and action frequency during work to determine work load and fatigue level.

[0081] 3. Data collection of psychological factors:

[0082] Psychological factors involve workers' mental health, work stress, motivation and morale, etc. This data can be collected through the following ways:

[0083] (1) Mental health scale: Use standardized mental health assessment tools such as GHQ (General Health Questionnaire) or PHQ-9 (Patient Health Questionnaire) to assess workers' mental health status, including anxiety, depression, etc.

[0084] (2) Pressure Assessment Questionnaire: Use scales such as the Perceived Stress Scale (PSS) to assess workers' stress levels at work, and combine physiological data (such as heart rate and skin conductance) to verify the questionnaire results;

[0085] (3) Motivation and Morale Interviews: Through in-depth interviews, understand the motivation level and attitude towards work of workers. The intrinsic and extrinsic motivation of workers can be evaluated using the questionnaire in Self-Determination Theory (SDT);

[0086] (4) Emotion Recognition Technology: Through facial expression analysis and voice emotion recognition technology, real-time monitoring of workers' emotional state to assess their mental health and emotional fluctuations on work.

[0087] 4. Skill Factor Data Collection:

[0088] Skill factors involve whether the skill level of workers matches the task requirements. This type of data collection can be done through the following methods:

[0089] (1) Skill Assessment Test: Design standardized skill assessment tests to examine workers' specific skill levels in the assembly process, such as part identification, tool use, fine operation, etc.

[0090] (2) Performance Record Analysis: Analyze workers' performance records in specific assembly tasks to collect data related to skill matching, and use time analysis to evaluate the time and accuracy required for workers to complete tasks;

[0091] (3) Training Record Collection: Collect workers' past skill training records to understand their skill background; evaluate performance after training to verify training effectiveness.

[0092] 5. Organizational Factor Data Collection:

[0093] Organizational factors involve rules and regulations, skill training, and management methods that affect workers' performance. This type of data collection method includes:

[0094] (1) Rules and Management Interviews: Interview management to understand the organization's rules and regulations, work processes and management strategies; analyze how these organizational factors affect workers' behavior and performance;

[0095] (2) Training Effectiveness Evaluation Questionnaire: After skill training, use a questionnaire to evaluate the effectiveness of the training, including the degree of mastery of new skills and their application in work;

[0096] (3) Employee Satisfaction Survey: Assess worker satisfaction with management style, work environment, and rules and regulations through a survey questionnaire. Employee satisfaction is an important indicator for evaluating the impact of organizational factors;

[0097] (4) Performance Data Analysis: Collect performance data such as worker productivity, work quality, and task completion time to quantify the impact of organizational factors on work performance.

[0098] After that, the embodiments of the present application can form the following specific human factor and corresponding data collection method library table, as shown in Table 1:

[0099] Table 1

[0100]

[0101]

[0102] Therefore, the final data collection method in the embodiments of the present application needs to be selected comprehensively considering factors such as data precision requirements, worker work impact, cost and feasibility.

[0103] In step S103, it is judged whether the start face condition of the complex product assembly sequence planning model is single start face condition. If the start face condition is single start face condition, the complex product assembly sequence planning model under single start face condition is solved based on the preset heuristic algorithm to obtain the corresponding target assembly sequence.

[0104] In step S104, if the start face condition is multi-start face condition, the assembly time, assembly cost, multiple important human factor factors and multiple constraint conditions of the complex product assembly sequence planning model under multi-start face condition are determined, and the fitness function is constructed according to the assembly time, assembly cost, multiple important human factor factors and multiple constraint conditions. The complex product assembly sequence planning model under multi-start face condition is solved by the fitness function and the preset hybrid genetic programming algorithm to obtain the corresponding target assembly sequence.

[0105] In the complex product assembly sequence planning, it is crucial to quantify and express human factors into operational optimization models. In order to comprehensively and accurately describe these factors, the following will describe the mathematical expression method of cognitive factors, physiological factors, psychological factors, skill factors, and organizational factors in five dimensions.

[0106] After the mathematical expression and mathematical modeling of human factors, for the complex product assembly sequence planning (ASP) problem model, the embodiments of the present application can consider the following two cases:

[0107] I. For the case of a single workface (i.e. there is only one worker on the assembly line to perform linear assembly tasks caozuo1), this kind of situation can use heuristic algorithms such as genetic algorithm to solve the problem;

[0108] II. For the problem of multiple workfaces (i.e. there are multiple workers on the assembly line to carry out assembly tasks in parallel).

[0109] Parallel assembly sequence planning is particularly necessary, so in the model solving link, the embodiment of the application proposes a special solving method for parallel workface design. In the traditional linear assembly mode, each assembly task depends on the completion of the previous task, often leading to production bottlenecks and time delays. Parallel assembly mode allows multiple assembly tasks to be performed simultaneously, which can effectively improve the flexibility and response speed of the production line. Especially in the face of diversified and personalized market demand, parallel assembly mode can better adapt to the rapidly changing production environment, thereby achieving higher production efficiency and shorter delivery cycle. In the actual assembly process of complex products, the required material resources are often over-provisioned, but the constraint of human resources is universal, that is, the number of assembly personnel available at each assembly workstation often determines the maximum workface number.

[0110] In solving the above assembly sequence planning problem, hybrid genetic programming (HGP) as a powerful optimization algorithm has good adaptability and global search ability. Compared with traditional heuristic algorithms, HGP combines the global search advantage of genetic algorithm with the flexibility of programming, which can effectively handle complex assembly sequence planning problems; by introducing the consideration of personnel constraints, HGP can not only optimize resource allocation in assembly tasks, but also ensure the reasonable use of human resources, thereby improving the overall assembly efficiency.

[0111] In order to solve the optimal assembly sequence planning problem under the current maximum workface constraint, the embodiment of the application adopts the method of genetic programming (GP). Genetic programming is an optimization algorithm based on evolutionary ideas, which is suitable for solving complex combinatorial optimization problems. Through genetic programming, the optimal or near-optimal assembly sequence can be found under the consideration of multiple constraints and objectives.

[0112] Optionally, in an embodiment of the present application, a plurality of important human factor data are used to mathematically model each important human factor to obtain a complex product assembly sequence planning model corresponding to a target complex product assembly field, including: quantifying worker attention, worker memory capacity and worker decision-making capacity in a complex product assembly task execution process of the target complex product assembly field to construct a cognitive factor mathematical model according to the quantified worker attention, worker memory capacity and worker decision-making capacity; obtaining a working time, a rest time and at least one physiological index of the worker in the complex product assembly task execution process, determining a fatigue degree of the worker according to the working time and the rest time, and estimating a physical state of the worker through the at least one physiological index to construct a physiological factor mathematical model according to the fatigue degree and the physical state; quantifying an emotional state of the worker in the complex product assembly task execution process through a pre-set emotional score scale, obtaining at least one physiological parameter of the worker, and quantifying a stress level of the worker using the at least one physiological parameter and a pre-set subjective report to construct a psychological factor mathematical model based on the quantified emotional state and stress level; obtaining an average time and a benchmark time for the worker to complete the same complex product assembly task, and calculating a skill level of the worker according to the average time and the benchmark time; obtaining a participation number of the worker in an assembly task that meets a pre-set similarity requirement with the complex product assembly task, and constructing a skill factor mathematical model according to the skill level and the participation number; determining a training degree, a rule compliance degree and a team cooperation efficiency of the worker, and constructing an organizational factor mathematical model according to the training degree, the rule compliance degree and the team cooperation efficiency.

[0113] It should be noted that the process of mathematically modeling the cognitive factor, the physiological factor, the psychological factor, the skill factor and the organizational factor in the embodiment of the present application is as follows:

[0114] 1. Mathematical expression of cognitive factor:

[0115] The cognitive factor includes worker attention, memory capacity, decision-making capacity and the like in the task execution process, which will affect the accuracy and efficiency of the worker in completing the assembly task; the embodiment of the present application can be quantified by the following methods:

[0116] 1. Attention level (A t ):

[0117] Attention can be reflected by the number of errors in a specific time period, the higher the attention level, the fewer the number of errors, which means that the worker's attention is more concentrated in the time period.

[0118] Let the attention level at time t be A t , which is inversely proportional to the number of errors E t , which can be expressed as:

[0119] At = 1 / (1 + k1*E t )

[0120] wherein (E t ) represents the number of errors occurred during time (t), (k1) is a weight factor for adjusting the degree of influence of errors on attention; the formula indicates that when the number of errors increases, the level of attention decreases;

[0121] 2. Decision-making ability (D c ):

[0122] Decision-making ability can be represented by the correct decision rate of workers when performing multiple tasks; the higher the decision-making ability, the more correct decisions workers can make in a multi-task environment.

[0123] D c = n correct / n total

[0124] wherein (n correct ) is the number of correct decisions, (n total ) is the total number of decisions, the larger the ratio, the stronger the decision-making ability of the worker.

[0125] 2. Mathematical expression of physiological factors:

[0126] Physiological factors mainly include fatigue level, physical condition, etc., which have a significant impact on the efficiency of workers completing tasks, as described below:

[0127] 1. Fatigue level (F t ):

[0128] Fatigue level can be measured by the length of work time and rest time of workers, fatigue level increases with the increase of work time and decreases with the increase of rest time.

[0129] F t = F (t-1) + k2*W t - k3*R t

[0130] wherein F (t-1) represents the fatigue level of the previous time period, W t represents the length of work time at time t, R t represents the length of rest time at time t, k2 and k3 are weight factors for fatigue growth and recovery respectively; the formula indicates that with the increase of work time, fatigue level will rise, while rest time helps to recover fatigue.

[0131] 2. Physical condition (P sPhysical condition can be estimated by indicators such as heart rate and blood pressure of workers, usually represented by a standardized physical fitness index:

[0132] P s = (HR max -HR rest ) / HR max

[0133] where HR max is the maximum heart rate, and HR rest is the resting heart rate. This formula represents the difference between physical condition and heart rate, the greater the difference, the better the worker's physical condition.

[0134] 3. Mathematical expression of psychological factors:

[0135] Psychological factors include emotional state, stress level, etc., which can affect the worker's concentration and completion quality in the task, as follows:

[0136] 1. Emotional state (M e ):

[0137] The application example can be quantified by the score measured by the emotional score scale, ranging from -1 (extremely negative) to 1 (extremely positive):

[0138] M e ∈ [-1, 1]

[0139] where the higher the value of M e indicates a more positive emotional state, and a negative value indicates a negative emotional state.

[0140] 2. Stress level (S t ):

[0141] Stress level can be quantified by subjective reports combined with physiological parameters such as skin conductance and heart rate variability:

[0142] S t = k4*HRV + k5*GSR

[0143] where HRV is heart rate variability, GSR is skin conductance, and k4 and k5 are weight factors used to adjust the influence of heart rate variability and skin conductance on stress level. Changes in heart rate variability and skin conductance can reflect the worker's psychological stress level.

[0144] 4. Mathematical expression of skill factors:

[0145] Skill factors include the worker's professional skill level, experience, etc., which have a direct impact on the efficiency and quality of the assembly task, as follows:

[0146] 1. Skill level (Sk ):

[0147] In the embodiments of the present application, the skill level can be represented according to the ratio of the average time of the worker to complete the same task to the benchmark time:

[0148] S k =T baseline / T worker

[0149] wherein T baseline is the benchmark time, representing the time to complete the task in an ideal case, and T worker is the average time of the worker to actually complete the task, the greater the value, the higher the skill level of the worker, and the more efficient the worker can complete the task.

[0150] 2, Experience (E x ):

[0151] Experience can be quantified by the number of times the worker participates in similar tasks.

[0152] E x =n task

[0153] wherein n task is the number of times the worker participates in similar tasks, the more times, the more experienced the worker is.

[0154] 5, Mathematical expression of organizational factors:

[0155] Organizational factors include training, rule compliance, team collaboration, etc., which will affect the overall efficiency and collaboration effect of the team, as described below:

[0156] 1, Training level (T r ):

[0157] The embodiments of the present application can be represented by the number of training courses participated by the worker and their grades:

[0158] T r =∑(C i *G i ), i=1,2,...,n

[0159] wherein C i is the weight of the i-th course, representing the importance of the course, G i is the grade of the worker in the i-th course, and n is the number of training courses; the formula is used to evaluate the overall training level of the worker.

[0160] 2, Rule compliance (C r ):

[0161] The rule compliance degree can be quantified by the number of violations of the worker in the work:

[0162] C r = 1 - (n vio / n checks )

[0163] where n vio is the number of violations, n checks is the total number of checks, the value is closer to 1, the higher the rule compliance degree, the less the number of violations.

[0164] 3、Team collaboration efficiency (C e ):

[0165] Team collaboration can be represented by the ratio of the time the team takes to complete the task to the ideal time:

[0166] C e = T iaeal / T actual

[0167] where T ideal is the ideal task completion time, indicating the time the team should complete the task in an ideal situation, T actual is the actual time the team takes to complete the task, the larger the value, the higher the team collaboration efficiency.

[0168] Optionally, in one embodiment of this application, the assembly time, assembly cost, multiple important human factors, and multiple constraints of the complex product assembly sequence planning model under multi-workface conditions are determined, and a fitness function is constructed based on the assembly time, assembly cost, multiple important human factors, and multiple constraints. The fitness function and a preset hybrid genetic programming algorithm are used to solve the complex product assembly sequence planning model under multi-workface conditions to obtain the corresponding target assembly sequence. This includes: reading and parsing multiple constraints of the parallel assembly sequence planning, wherein the multiple constraints include assembly process constraints, resource constraints, space constraints, maximum number of workfaces, and human factor constraints; determining the factor score and weight coefficient of each important human factor corresponding to each complex product assembly task, and constructing a human factor evaluation function based on the factor score and weight coefficient to evaluate the human factors based on the assembly time, assembly cost, and human factor constraints. A fitness function is constructed using an estimation function and multiple constraints. This fitness function determines the expression form of the assembly sequence planning result, and a neighborhood search operation is performed on the assembly sequence planning result to obtain the corresponding initial population. Based on a hybrid genetic programming algorithm, multiple individuals that meet the preset fitness requirements are selected from the initial population. According to a preset crossover strategy, each of these individuals is crossovered with its corresponding parent individual in the initial population to obtain multiple new offspring individuals. Based on a preset mutation probability, mutation operations are performed on these new offspring individuals, and the fitness of these mutated offspring individuals is evaluated to obtain the fitness function value corresponding to each new offspring individual. The selection, crossover, mutation, and fitness evaluation operations are iteratively executed until a preset termination condition is met to obtain the target assembly sequence that meets the preset fitness value requirements.

[0169] Specifically, the embodiments of this application first need to read and parse various constraints of the parallel assembly sequence planning. These constraints mainly include assembly process constraints, resource constraints, space constraints, and human factor constraints. Among them, the maximum number of working faces is a key limiting factor, representing the maximum number of assembly operations that can be performed simultaneously.

[0170] In the embodiments of this application, the assembly task set is set as A = {a1, a2, ..., a...} N} where N is the total number of assembly tasks; each assembly task has prerequisite and subsequent tasks, which constitute the sequence of assembly processes; the process constraints between assembly tasks can be represented as a set of relationships:

[0171] a j ∈A, a i <a j Indicates a i Must be in a j Completed previously

[0172] Resource constraints include the finiteness of human resources, equipment resources, etc., and need to ensure that the allocated resources do not exceed the total amount available at any time; spatial constraints relate to the spatial limitations of the assembly site, and ensure that the number of open work surfaces does not exceed the maximum limit P in the same time period:

[0173]

[0174] wherein δ k，t represents whether task a k is executed at time t (δ k，t =1 represents execution, and 0 represents non-execution), and T is the entire assembly time range.

[0175] Secondly, in order to evaluate the pros and cons of different assembly sequences, the embodiments of the present application also need to establish a fitness function that comprehensively considers assembly time, assembly cost, human factors, and the degree of satisfaction of the constraint conditions.

[0176] Optionally, in an embodiment of the present application, the mathematical expression of the fitness function is:

[0177] Fitness(X) = θ·T total (X) + β·C total (X) + γ·F human (X) + λ·Penalty(X)

[0178] wherein X represents an assembly sequence scheme; T total (X) represents the assembly time of the assembly sequence scheme X; C total (X) represents the assembly cost; F human (X) represents the human factor evaluation function; Penalty(X) represents the penalty term for violating the constraint condition; θ represents the weight coefficient corresponding to the assembly time; β represents the weight coefficient corresponding to the assembly cost; γ represents the weight coefficient corresponding to the human factor evaluation function; and λ represents the weight coefficient corresponding to the penalty term.

[0179] In the embodiments of the present application, the fitness function for evaluating the assembly sequence is as shown in the following formula:

[0180] Fitness(X) = θ·T total (X) + β·C total (X) + γ·F human (X) + λ·Penalty(X)

[0181] wherein x represents an assembly sequence scheme; T total (X) is the total assembly time of the assembly sequence scheme X; C total (X) is the total assembly cost; and F human(X) is an evaluation function of human factors, measuring the influence of the assembly sequence on workers; Penalty(X) is a penalty term for violating the constraint condition; θ, β, γ, λ are weight coefficients for balancing the importance of each index.

[0182] It should be noted that the total assembly time T total (X) can be calculated by scheduling simulation of the assembly sequence, and the latest time point at which all tasks are completed; the total assembly cost C total (X) includes labor cost, equipment usage cost, etc.

[0183] The human factor evaluation function F human (X) can be defined as:

[0184]

[0185] Wherein, H cog (a i ) is the cognitive factor score of task a i ; H phys (a i ) is the physiological factor score of task a i ; H psych (a i ) is the psychological factor score of task a i ; H skill (a i ) is the skill factor score of task a i ; H org (a i ) is the organizational factor score of task a i ; w cog , w phys , w psych , w skill , w org are weight coefficients corresponding to the factors.

[0186] In actual execution, the embodiments of the present application can adopt the following methods for corresponding mathematical processing of human factors:

[0187] 1. Cognitive factor score H cog (a i ):

[0188] Considering the cognitive load of workers when performing task a i and the understanding degree of the assembly process, the embodiments of the present application can evaluate through task complexity C task (a i ) and worker experience level E(a i ):

[0189]

[0190] Among them, the higher the task complexity and the lower the experience level, the higher the cognitive load score.

[0191] 2. Physiological Factors Score H phys (a i ):

[0192] The embodiments of this application take into account the worker's fatigue state, workload, and physiological comfort, and are based on the worker's cumulative working time t. work (a i ) and rest time t rest (a i Calculate physiological factor scores:

[0193] H phys (a i )=φ·(t work (a i )-t rest (a i ))

[0194] Where φ is the physiological load coefficient.

[0195] 3. Psychological Factors Score H psych (a i ):

[0196] Embodiments of this application can evaluate a worker's performance in performing task a i Stress, motivation, and satisfaction during the period can be measured by the psychological stress index P(a). i ) and motivation level M(a i ) Measure psychological factors score:

[0197] H psych (a i )=P(a i )-θ·M(a i )

[0198] Where θ is the motivation influence coefficient.

[0199] 4. Skills Factor Score H skill (a i ):

[0200] Embodiments of this application can measure the degree of matching between worker skills and the complexity of assembly tasks, and use the skill matching degree S(a i Calculate skill factor scores:

[0201]

[0202] A high score indicates a negative impact on adaptability when the skill matching degree is low.

[0203] 5. Organizational factor score H org (a i ):

[0204] Embodiments of the present application consider the influence of organizational regulations, skill training and management methods on workers' performance of tasks a i , and can evaluate the organizational factor score through organizational support O(a i ):

[0205]

[0206] The higher the organizational support, the lower the score, and the less negative impact on fitness.

[0207] In addition, the penalty term Penalty(X) in embodiments of the present application can be used to impose a penalty on the assembly sequence that violates the constraint condition, to guide the genetic programming to search in the feasible solution space. The penalty term can be defined as the total degree of violation of the constraint condition:

[0208]

[0209] where M is a large positive number, and Δ 资源 and Δ 空间 represent the degree to which the resource and space constraints are exceeded, respectively.

[0210] Thus, embodiments of the present application can more comprehensively evaluate the pros and cons of the assembly sequence by designing a fitness function that comprehensively considers assembly time, assembly cost, human factors and constraint conditions, thereby improving the quality of the optimization result.

[0211] Again, embodiments of the present application can determine the expression form of the assembly sequence planning result. Specifically, the result of assembly sequence planning needs to be expressed in a clear and easy-to-operate form. In embodiments of the present application, a tree structure can be used to represent the assembly sequence, with each node representing an assembly task and the connection between nodes representing the sequence of tasks.

[0212] Embodiments of the present application set the assembly sequence representation as a directed acyclic graph (DAG), denoted as:

[0213] G=(V,E),

[0214] where V is the node set, i.e. the assembly task, and E is the directed edge set, representing the sequence of tasks.

[0215] It can be understood that the above representation not only clearly expresses the assembly process flow, but also facilitates operations such as crossover and mutation in genetic programming.

[0216] Afterwards, embodiments of the present application can use a neighborhood search module to initialize the solution.

[0217] Specifically, in order to improve the search efficiency of genetic programming, embodiments of the present application can first use a neighborhood search module to initialize the solution; neighborhood search is an optimization method based on local search, which finds a better solution by adjusting the initial solution in a small range.

[0218] The generation of the initial solution can be based on the topological sorting of the assembly task. Under the premise of meeting the process constraints, a number of feasible assembly sequences are randomly generated as the initial population, which needs to cover different areas of the solution space to increase diversity.

[0219] In neighborhood search, for each initial solution X0, its neighborhood solution set N(X0) is generated through operations such as exchange, insertion, deletion, etc.; then, a solution with higher fitness is selected from the neighborhood solution as the initial population of genetic programming.

[0220] Specifically, the neighborhood search operation includes the following steps:

[0221] Step 1, exchange operation: exchange the positions of two non-directly related tasks in the assembly sequence;

[0222] Step 2, insertion operation: insert a task into another position, provided that it does not violate the process constraints;

[0223] Step 3, reverse operation: reverse a subsequence to change its execution order.

[0224] Thus, embodiments of the present application can obtain a batch of high-quality feasible solutions in the initial stage through neighborhood search, thereby laying a foundation for subsequent optimization of genetic programming.

[0225] Then, after obtaining the initial population, embodiments of the present application can use genetic programming method to solve the problem, i.e., using genetic programming method to optimize the assembly sequence, which mainly includes selection, crossover, mutation and fitness evaluation processes, as described below:

[0226] 1. Selection operation:

[0227] In embodiments of the present application, the selection operation aims to select individuals with higher fitness from the current population to enter the next generation, which can use methods such as roulette selection, tournament selection, etc. Assuming the population size is Np, the fitness value is Fitness(X i ), then the selection probability pi can be defined as:

[0228]

[0229] where N p is the population size, Fitness(X i ) is the fitness value of individual X i .

[0230] It should be noted that this method can ensure that individuals with higher fitness (i.e., smaller Fitness(X i )) have a greater probability of being selected.

[0231] 2. Crossover operation:

[0232] The crossover operation in the embodiments of the present application can generate new offspring individuals by exchanging part of the structure of two parent individuals; since the assembly sequence is represented in the form of a tree or a graph, the crossover operation needs to ensure that the generated offspring still meet the process constraints; the specific process of the crossover method used in the embodiments of the present application is as follows:

[0233] (1) Randomly select two parent individuals X parent1 and X parent2 ;

[0234] (2) Randomly select a subtree (subgraph) in each assembly sequence;

[0235] (3) Exchange the two subtrees to generate two new offspring individuals X child1 and X child2 ;

[0236] (4) Check whether the offspring individuals meet the constraint conditions, and if not, repair or discard them.

[0237] 3. Mutation operation:

[0238] In the embodiments of the present application, the mutation operation makes minor random modifications to individuals to increase the diversity of the population; this mutation operation needs to be carefully designed to avoid damaging the feasibility of the assembly sequence; the specific process of this mutation operation is as follows:

[0239] (1) Node replacement: replace a certain assembly task with another alternative task;

[0240] (2) Subtree recombination: rearrange a certain subtree to change the order of its internal tasks;

[0241] (3) Task insertion / deletion: insert or delete a certain assembly task without violating the process constraints.

[0242] In the specific implementation process, the mutation probability is usually set low to maintain the stability of the population.

[0243] 4. Fitness evaluation:

[0244] The embodiment of the present application can calculate the fitness function value Fitness(X) of the newly generated offspring individual, and if the offspring individual violates the constraint condition, the fitness value will become worse due to the existence of the penalty term, so that the offspring individual will be eliminated in the selection process.

[0245] 5. Iteration operation:

[0246] The embodiment of the present application can repeat the above selection, crossover, mutation and fitness evaluation process until the termination condition is met, which can be reaching the preset maximum iteration number G max , or the fitness value is not significantly improved in a plurality of consecutive generations.

[0247] It should be noted that in the embodiment of the present application, the goal of genetic programming is to solve the following optimization problem:

[0248]

[0249] Wherein, is the set of all feasible assembly sequences.

[0250] Further, the embodiment of the present application can summarize the detailed steps of each step above to obtain a general flow of using genetic programming algorithm to solve the assembly sequence planning problem:

[0251] 1. Initialization: generating an initial population using neighborhood search

[0252] 2. Fitness evaluation: calculating the fitness value of each individual in the initial population;

[0253] 3. Iteration loop (for g = 1 to G max ):

[0254] (1) Selection: selecting individuals from the population according to the fitness value;

[0255] (2) Crossover: performing crossover operation on the selected individuals to generate offspring individuals;

[0256] (3) Mutation: performing mutation operation on the offspring individuals;

[0257] (4) Fitness evaluation: calculating the fitness value of the new individual;

[0258] (5) Population update: generating a new population

[0259] 5. Output: outputting the assembly sequence with the optimal fitness value after the termination condition is met.

[0260] In the assembly sequence planning, the embodiments of the present application need to ensure that the number of simultaneous assembly tasks does not exceed the maximum open face limit P; in the fitness evaluation and constraint checking of genetic programming, the number of assembly tasks in progress needs to be counted at each time point t:

[0261]

[0262] wherein δ i，t represents whether the task a i is in the progress state at time t.

[0263] It can be understood that the embodiments of the present application use genetic programming method to solve the assembly sequence, which has the advantages of strong global search ability and high adaptability, so as to find the optimal or near-optimal solution under the consideration of multiple constraints and targets.

[0264] Optionally, in an embodiment of the present application, after solving the complex product assembly sequence planning model under the multi-open face condition by using the fitness function and the preset hybrid genetic programming algorithm to obtain the corresponding target assembly sequence, the method further comprises: acquiring current human factor data of workers in the process of performing the complex product assembly task; and dynamically adjusting a human factor evaluation function in the fitness function according to the current human factor data, so as to solve the complex product assembly sequence planning model under the multi-open face condition by using the dynamically adjusted human factor evaluation function.

[0265] In addition, since the human factor is dynamic in the assembly process, for example, the fatigue degree of the worker will increase with time. Therefore, the embodiments of the present application can obtain the state data of the worker in real time by using digital thread and sensor technology, so that in the process of genetic programming, the F human (X) in the fitness function can be dynamically adjusted according to the latest human factor data.

[0266] As an implementable way, the genetic programming method in the embodiments of the present application can be programmed and implemented in the python tool, and the corresponding effect can be obtained according to the input number of workers / maximum open face limit, as shown in Figure 2 and Figure 3 In the specific implementation process, the implemented example uses an assembly body with 23 assembly tasks to demonstrate, Figure 2 which shows the case of only one worker, and a linear assembly task sequence is obtained; Figure 3 which shows the case that 5 workers can simultaneously start assembly, and the algorithm result output is in the form of hierarchical display of assembly tasks contained in each open face.

[0267] In summary, the embodiment of the application introduces the cognitive, physiological, psychological, skill and organizational human factors in the complex product assembly sequence planning process, optimizes and solves the assembly sequence based on genetic programming, acquires the state data of workers through effective decision-making human factor data acquisition mode, adjusts the assembly sequence in combination with the constraint of human factors, and finally realizes the optimal assembly sequence planning considering the state of workers, so as to improve the efficiency and reliability of the assembly process.

[0268] The solving method of the complex product assembly sequence planning problem considering human factors according to the embodiment of the application considers the human factors in the assembly process, including cognitive, physiological, psychological, skill and organizational factors, and comprehensively uses the genetic programming technology to find the optimal or approximate optimal assembly sequence, so as to ensure efficient assembly under the maximum opening face limit; in addition, the embodiment covers the whole process method of identification, data collection and analysis, operation optimization modeling of key human factors in the complex product assembly scene, and solving using genetic programming, and provides an efficient and reliable assembly sequence planning solution for complex product systems.

[0269] Secondly, the solving device of the complex product assembly sequence planning problem considering human factors according to the embodiment of the application is described with reference to the accompanying drawings.

[0270] Figure 4 is the block schematic diagram of the solving device of the complex product assembly sequence planning problem considering human factors according to the embodiment of the application.

[0271] As Figure 4 shown, the solving device of the complex product assembly sequence planning problem considering human factors 10 includes a modeling module 100, an acquisition module 200, a first solving module 300 and a second solving module 400.

[0272] The modeling module 100 is configured to establish a human factor library in the target complex product assembly field based on a preset human factor identification framework, construct a priority evaluation model according to a preset analytic hierarchy process strategy, and calculate the comprehensive score of each human factor in the human factor library by using the priority evaluation model, wherein the human factor library includes cognitive factors, physiological factors, psychological factors, skill factors and organizational factors.

[0273] The acquisition module 200 is configured to acquire human factor data corresponding to each human factor based on a pre-constructed data acquisition mode library, and perform mathematical modeling on each human factor by using the human factor data, so as to obtain a complex product assembly sequence planning model corresponding to the target complex product assembly field.

[0274] The first solving module 300 is configured to determine the start face condition of the complex product assembly sequence planning model, and if the start face condition is a single start face condition, solve the complex product assembly sequence planning model under the single start face condition based on a preset heuristic algorithm to obtain a corresponding target assembly sequence.

[0275] The second solving module 400 is configured to determine the assembly time, assembly cost, each human factor and multiple constraint conditions of the complex product assembly sequence planning model under the multiple start face condition, and construct a fitness function according to the assembly time, assembly cost, each human factor and multiple constraint conditions, so as to solve the complex product assembly sequence planning model under the multiple start face condition by comprehensively scoring, the fitness function and a preset hybrid genetic programming algorithm to obtain a corresponding target assembly sequence.

[0276] Optionally, in an embodiment of the present application, the modeling module 100 comprises a first constructing unit, a scoring unit, a consistency checking unit and a normalization unit.

[0277] The first constructing unit is configured to construct a human factor identification framework based on actual work requirements of a target complex product assembly field and a plurality of preset factor identification theories, and establish a human factor library of the target complex product assembly field according to the human factor identification framework, wherein the plurality of factor identification theories comprise a human factors engineering theory, a work design theory, a SHEL model, an operation risk management theory and an organizational behavior theory.

[0278] The scoring unit is configured to score the relative importance of each pair of human factors in the human factor library to construct a judgment matrix, and calculate the maximum eigenvalue and consistency index of the judgment matrix.

[0279] The consistency checking unit is configured to calculate a consistency ratio based on the maximum eigenvalue and consistency index, and perform consistency checking on each pair of human factors according to the consistency ratio to obtain a corresponding consistency checking result.

[0280] The normalization unit is configured to normalize each column of data of the judgment matrix to calculate the weight of each human factor, and determine the comprehensive score of each human factor in the human factor library based on the weight and consistency checking result.

[0281] Optionally, in an embodiment of the present application, the collecting module 200 comprises a determining unit and a second constructing unit.

[0282] The determining unit is configured to determine the cognitive factor data collection mode, physiological factor data collection mode, psychological factor data collection mode, skill factor data collection mode and organizational factor data collection mode corresponding to the cognitive factor, physiological factor, psychological factor, skill factor and organizational factor, respectively.

[0283] a second construction unit, configured to construct a data collection mode library according to the cognitive factor data collection mode, the physiological factor data collection mode, the psychological factor data collection mode, the skill factor data collection mode and the organizational factor data collection mode.

[0284] Optionally, in an embodiment of the present application, the collection module 200 further comprises a quantification unit, an estimation unit, a first acquisition unit, a calculation unit, a second acquisition unit and a third construction unit.

[0285] The quantification unit is configured to quantify the worker attention, the worker memory and the worker decision-making ability in the complex product assembly task execution process of the target complex product assembly field, and to construct a cognitive factor mathematical model according to the quantified worker attention, worker memory and worker decision-making ability.

[0286] The estimation unit is configured to acquire the worker working time, the worker resting time and at least one physiological index in the complex product assembly task execution process, to determine the worker fatigue degree according to the working time and the resting time, and to estimate the worker physical state through the at least one physiological index, so as to construct a physiological factor mathematical model according to the fatigue degree and the physical state.

[0287] The first acquisition unit is configured to quantify the worker emotional state in the complex product assembly task execution process through a preset emotional score scale, to acquire at least one physiological parameter of the worker, and to quantify the worker stress level by using the at least one physiological parameter and a preset subjective report, so as to construct a psychological factor mathematical model based on the quantified emotional state and stress level.

[0288] The calculation unit is configured to acquire the average time and the reference time of the worker to complete the same complex product assembly task, and to calculate the skill level of the worker according to the average time and the reference time.

[0289] The second acquisition unit is configured to acquire the participation frequency of the worker in the assembly task that meets the preset similarity requirement with the complex product assembly task, and to construct a skill factor mathematical model according to the skill level and the participation frequency.

[0290] The third construction unit is configured to determine the training degree, the rule compliance degree and the team cooperation efficiency of the worker, and to construct an organizational factor mathematical model according to the training degree, the rule compliance degree and the team cooperation efficiency.

[0291] Optionally, in an embodiment of the present application, the second solving module 400 comprises a reading unit, a neighborhood search unit, an algorithm processing unit and an iteration unit.

[0292] The reading unit is configured to read and analyze a plurality of constraint conditions of the parallel assembly sequence planning, wherein the plurality of constraint conditions comprise assembly process constraints, resource constraints, space constraints, maximum open face quantity constraints and human factor constraints.

[0293] The neighborhood searching unit is configured to determine a factor score and a weight coefficient of each important human factor corresponding to each complex product assembly task, and construct a human factor evaluation function based on the factor score and the weight coefficient, so as to construct a fitness function based on the assembly time, the assembly cost, the human factor evaluation function and the plurality of constraint conditions, and determine an expression form of the assembly sequence planning result based on the fitness function, and perform a neighborhood search operation on the assembly sequence planning result to obtain a corresponding initial population.

[0294] The algorithm processing unit is configured to filter a plurality of individuals satisfying a preset fitness requirement from the initial population based on the hybrid genetic programming algorithm, and perform a crossover operation on each individual in the plurality of individuals and a parent individual corresponding to the individual in the initial population according to a preset crossover strategy to obtain a plurality of new child individuals.

[0295] The iteration unit is configured to perform a mutation operation on the plurality of new child individuals based on a preset mutation probability, and perform fitness evaluation on the plurality of new child individuals after the mutation operation to obtain a fitness function value corresponding to each new child individual in the plurality of new child individuals, and iteratively perform the filtering, the crossover, the mutation and the fitness evaluation operations until a preset termination condition is satisfied, so as to obtain a target assembly sequence satisfying a preset fitness value requirement.

[0296] Optionally, in an embodiment of the present application, the complex product assembly sequence planning problem solving device 10 considering human factors in the embodiment of the present application further comprises a data acquisition module and a dynamic adjustment module.

[0297] The data acquisition module is configured to acquire current human factor data of workers in the process of performing complex product assembly tasks based on a data acquisition mode library after solving a complex product assembly sequence planning model in a multi-open face situation based on the comprehensive score, the fitness function and the preset hybrid genetic programming algorithm to obtain a corresponding target assembly sequence.

[0298] The dynamic adjustment module is configured to dynamically adjust a human factor evaluation function in the fitness function according to the current human factor data, so as to solve the complex product assembly sequence planning model in the multi-open face situation by using the dynamically adjusted human factor evaluation function.

[0299] Optionally, in an embodiment of the present application, the mathematical expression of the fitness function is:

[0300] Fitness(X)=θ·T total (X)+β·Ctotal (X) + γ · F human (X) + λ · Penalty(X)

[0301] Wherein, X represents assembly sequence scheme; T total (X) represents assembly time of assembly sequence scheme X; C total (X) represents assembly cost; F human (X) represents human factor evaluation function; Penalty(X) represents penalty term of constraint condition violation; θ represents weight coefficient corresponding to assembly time; β represents weight coefficient corresponding to assembly cost; γ represents weight coefficient corresponding to human factor evaluation function; λ represents weight coefficient corresponding to penalty term.

[0302] It should be noted that the aforementioned explanation and description of the embodiment of the method for solving the complex product assembly sequence planning problem considering human factors is also applicable to the device for solving the complex product assembly sequence planning problem considering human factors of the embodiment, which will not be described here.

[0303] The device for solving the complex product assembly sequence planning problem considering human factors provided by the embodiment of the application comprises a modeling module, which is configured to establish a human factor library in a target complex product assembly field based on a preset human factor identification framework, and to construct a priority evaluation model according to a preset analytic hierarchy process strategy, and to calculate a comprehensive score of each human factor in the human factor library by using the priority evaluation model, wherein the human factor library comprises cognitive factors, physiological factors, psychological factors, skill factors and organizational factors; a collection module, which is configured to collect human factor data corresponding to each human factor based on a pre-constructed data collection mode library, and to perform mathematical modeling on each human factor by using the human factor data, so as to obtain a complex product assembly sequence planning model corresponding to the target complex product assembly field; a first solving module, which is configured to solve the complex product assembly sequence planning model in a single starting face case based on a preset heuristic algorithm, so as to obtain a corresponding target assembly sequence; and a second solving module, which is configured to determine assembly time, assembly cost, each human factor and a plurality of constraint conditions of the complex product assembly sequence planning model in a multi-starting face case, and to construct a fitness function according to the assembly time, the assembly cost, each human factor and the plurality of constraint conditions, so as to solve the complex product assembly sequence planning model in the multi-starting face case by using the comprehensive score, the fitness function and a preset hybrid genetic programming algorithm, so as to obtain a corresponding target assembly sequence. The assembly sequence planning method of the application can efficiently and comprehensively consider human factors, and can realize planning of an optimal assembly sequence under complex constraint conditions, thereby improving assembly efficiency, reducing cost, and ensuring assembly quality and worker safety.

[0304] Figure 5 A structural schematic diagram of an electronic device is provided for the embodiment of the application. The electronic device can comprise:

[0305] The memory 501, the processor 502 and the computer program stored in the memory 501 and executable on the processor 502.

[0306] The processor 502 implements the method for solving the human-factor-considered complex product assembly sequence planning problem provided in the above embodiments when executing the program.

[0307] Further, the electronic device further comprises:

[0308] The communication interface 503 is used for communication between the memory 501 and the processor 502.

[0309] The memory 501 is used for storing the computer program executable on the processor 502.

[0310] The memory 501 can include a high-speed RAM memory, and can also include a non-volatile memory, for example, at least one disk memory.

[0311] If the memory 501, the processor 502 and the communication interface 503 are independently implemented, the communication interface 503, the memory 501 and the processor 502 can be connected to each other through a bus and complete communication between each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 5 In the figure, only one thick line is used to represent, but it does not mean that there is only one bus or one type of bus.

[0312] Optionally, in a specific implementation, if the memory 501, the processor 502 and the communication interface 503 are integrated on a chip, the memory 501, the processor 502 and the communication interface 503 can complete communication between each other through an internal interface.

[0313] The processor 502 can be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0314] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the above-mentioned method for solving a complex product assembly sequence planning problem considering human factors.

[0315] In the description of the present specification, the description referring to the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or N embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples, without contradiction.

[0316] In addition, the terms "first", "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "N" is at least two, for example, two, three, etc., unless otherwise specifically limited.

[0317] Any process or method descriptions in flow charts or otherwise described herein can be understood as representing code modules, segments, or portions of code that include one or more executable instructions for performing a specified logic function or process, and the various embodiments of the present application also include the possibility that the functions described can be implemented using a plurality of separate program components or objects to perform the described functions, and that these components or objects can be written in accordance with the present application and can be implemented with hardware that is specifically constructed to store and perform the executable instructions, or alternatively can be implemented with a general purpose receiver or other programmable computing device that can store and execute the software on an as needed basis.

[0318] The logic and / or steps represented in the flowcharts and / or described herein, for example, can be considered as a sequence of instructions to implement logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, processor- based system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For purposes of this specification, a "computer-readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be a computer- readable storage medium or a computer-readable signal medium. The computer- readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include the following: an electrical connection having one or more wires (electrical connections), a portable computer diskette (a magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via, for example, optical scanning of the paper or other medium, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

[0319] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented in hardware and in another embodiment, any of the following technologies, known in the art, or their combinations can be used: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0320] Those skilled in the art can understand that all or part of the steps carried out by the above-mentioned embodiments can be completed by programs instructing related hardware, and the programs can be stored in a computer-readable storage medium. When the programs are executed, one or a combination of the steps of the method embodiments is included.

[0321] In addition, each of the functional units in the various embodiments of the present application can be integrated in one processing module, or each of the units can be physically present separately, or two or more units can be integrated in one module. The integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer readable storage medium.

[0322] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

Claims

1. A method for solving a complex product assembly sequence planning problem considering human factors, characterized in that, The method comprises the following steps: Based on a preset human factor identification framework, a human factor library for a target complex product assembly field is established, and a priority evaluation model is constructed according to a preset analytic hierarchy process strategy, and the priority evaluation model is used to calculate the comprehensive score of each human factor in the human factor library, wherein the human factor library comprises cognitive factors, physiological factors, psychological factors, skill factors and organizational factors; Based on a pre-constructed data acquisition method library, human factor data corresponding to each human factor is acquired, and mathematical modeling is performed on each human factor by using the human factor data to obtain a complex product assembly sequence planning model corresponding to the target complex product assembly field; If the start face condition of the complex product assembly sequence planning model is a single start face condition, a corresponding target assembly sequence is obtained by solving the complex product assembly sequence planning model under the single start face condition based on a preset heuristic algorithm; If the start face condition is a multi-start face condition, the assembly time, assembly cost, each human factor and multiple constraint conditions of the complex product assembly sequence planning model under the multi-start face condition are determined, and a fitness function is constructed according to the assembly time, assembly cost, each human factor and multiple constraint conditions, so as to solve the complex product assembly sequence planning model under the multi-start face condition by using the comprehensive score, the fitness function and a preset hybrid genetic programming algorithm, and obtain a corresponding target assembly sequence; The mathematical modeling of each human factor by using the human factor data to obtain a complex product assembly sequence planning model corresponding to the target complex product assembly field comprises: Quantifying the worker's attention, memory capacity and decision-making ability in the complex product assembly task execution process of the target complex product assembly field to construct a cognitive factor mathematical model according to the quantified worker's attention, memory capacity and decision-making ability; Obtaining the working time, rest time and at least one physiological indicator of the worker in the complex product assembly task execution process, determining the worker's fatigue degree according to the working time and the rest time, and estimating the worker's physical state through the at least one physiological indicator, to construct a physiological factor mathematical model according to the fatigue degree and the physical state; Quantifying the emotional state of the worker in the complex product assembly task execution process by using a preset emotional score scale, obtaining at least one physiological parameter of the worker, and quantifying the stress level of the worker by using the at least one physiological parameter and a preset subjective report, to construct a psychological factor mathematical model based on the quantified emotional state and stress level; Obtaining the average time and the baseline time of the worker to complete the same complex product assembly task, and calculating the skill level of the worker according to the average time and the baseline time; acquire a participation number of an assembly task meeting a preset similarity requirement of the worker participation and the assembly task of the complex product, and construct a skill factor mathematical model according to the skill level and the participation number; determine a training degree, a rule compliance degree and a team cooperation efficiency of the worker, and construct an organization factor mathematical model according to the training degree, the rule compliance degree and the team cooperation efficiency; determine the assembly time, the assembly cost, each human factor and multiple constraint conditions of the complex product assembly sequence planning model in the multi-opening work surface condition, and construct an adaptive function according to the assembly time, the assembly cost, each human factor and the multiple constraint conditions, so as to solve the complex product assembly sequence planning model in the multi-opening work surface condition by the comprehensive score, the adaptive function and a preset hybrid genetic programming algorithm, to obtain a corresponding target assembly sequence, including: reading and analyzing multiple constraint conditions of parallel assembly sequence planning, wherein the multiple constraint conditions include assembly process constraints, resource constraints, space constraints, maximum opening work surface number and human factor constraints; determining a factor score and a weight coefficient of each important human factor corresponding to each complex product assembly task, and constructing a human factor evaluation function through the factor score and the weight coefficient, so as to construct the adaptive function according to the assembly time, the assembly cost, the human factor evaluation function and the multiple constraint conditions, and determine an expression form of the assembly sequence planning result through the adaptive function, and perform a neighborhood search operation on the assembly sequence planning result, to obtain a corresponding initial population; based on the hybrid genetic programming algorithm, screening multiple individuals meeting a preset adaptive requirement from the initial population, and performing a crossover operation on each individual in the multiple individuals and a parent individual corresponding to the individual in the initial population according to a preset crossover strategy, to obtain multiple new child individuals; based on a preset mutation probability, performing a mutation operation on the multiple new child individuals, and performing an adaptive evaluation on the multiple new child individuals after the mutation operation, to obtain an adaptive function value corresponding to each new child individual in the multiple new child individuals, and iteratively performing the screening, the crossover, the mutation and the adaptive evaluation operations until a preset termination condition is met, to obtain a target assembly sequence meeting a preset adaptive value requirement.

2. The method of claim 1, wherein, based on the preset human factor identification framework, establishing a human factor library of a target complex product assembly field, and constructing a priority evaluation model according to a preset analytic hierarchy process strategy, and calculating a comprehensive score of each human factor in the human factor library by using the priority evaluation model, including: based on actual work requirements of the target complex product assembly field and multiple factor identification theories, constructing the human factor identification framework, to establish the human factor library of the target complex product assembly field according to the human factor identification framework, wherein the multiple factor identification theories include a human factors engineering theory, a work design theory, a SHEL model, an operation risk management theory and an organizational behavior theory; The relative importance of each pair of human factors in the human factor library is scored to build a judgment matrix, and the maximum eigenvalue and consistency index of the judgment matrix are calculated; Based on the maximum eigenvalue and the consistency index, a consistency ratio is calculated, and a consistency test is performed on each pair of human factors according to the consistency ratio to obtain a corresponding consistency test result; Each column of data in the judgment matrix is normalized to calculate the weight of each human factor, and based on the weight and the consistency test result, the comprehensive score of each human factor in the human factor library is determined.

3. The method of claim 1, wherein, The human factor data corresponding to each human factor is collected based on the pre-constructed data collection method library, which includes: Respective cognitive factor data collection methods, physiological factor data collection methods, psychological factor data collection methods, skill factor data collection methods, and organizational factor data collection methods corresponding to the cognitive factor, the physiological factor, the psychological factor, the skill factor, and the organizational factor are determined; The data collection method library is constructed according to the cognitive factor data collection method, the physiological factor data collection method, the psychological factor data collection method, the skill factor data collection method, and the organizational factor data collection method.

4. The method of claim 1, wherein, After solving the complex product assembly sequence planning model in the multi-opening surface situation by the comprehensive score, the fitness function, and the preset hybrid genetic programming algorithm to obtain the corresponding target assembly sequence, it further includes: Based on the data collection method library, the current human factor data of workers in the process of performing the complex product assembly task is obtained; According to the current human factor data, the human factor evaluation function in the fitness function is dynamically adjusted to solve the complex product assembly sequence planning model in the multi-opening surface situation by using the dynamically adjusted human factor evaluation function.

5. The method of claim 1, wherein, The mathematical expression of the fitness function includes: wherein, denotes an assembly sequence scheme; denotes an assembly sequence scheme X denotes an assembly time of denotes an assembly cost; denotes a human factor evaluation function; denotes a penalty term for violating a constraint condition; denotes a weight coefficient corresponding to the assembly time; denotes a weight coefficient corresponding to the assembly cost; denotes a weight coefficient corresponding to the human factor evaluation function; denotes a weight coefficient corresponding to the penalty term.

6. A human factor consideration complex product assembly sequence planning problem solving apparatus for implementing the human factor consideration complex product assembly sequence planning problem solving method according to any one of claims 1 to 5, characterized in that, It includes: A modeling module is used to build a human factor library in a target complex product assembly field based on a preset human factor identification framework, and to build a priority evaluation model according to a preset analytic hierarchy process strategy, and to calculate the comprehensive score of each human factor in the human factor library using the priority evaluation model, wherein the human factor library includes cognitive factors, physiological factors, psychological factors, skill factors, and organizational factors; A collection module is used to collect human factor data corresponding to each human factor based on a pre-constructed data collection method library, and to mathematically model each human factor using the human factor data to obtain a complex product assembly sequence planning model corresponding to the target complex product assembly field; A first solving module is used to determine the opening surface situation of the complex product assembly sequence planning model, and if the opening surface situation is a single opening surface situation, to solve the complex product assembly sequence planning model in the single opening surface situation based on a preset heuristic algorithm to obtain a corresponding target assembly sequence. The second solving module is configured to determine assembly time, assembly cost, each human factor and multiple constraint conditions of the complex product assembly sequence planning model in the multi-open-face condition if the open-face condition is the multi-open-face condition, and construct a fitness function according to the assembly time, the assembly cost, the each human factor and the multiple constraint conditions, so as to solve the complex product assembly sequence planning model in the multi-open-face condition by the comprehensive score, the fitness function and a preset hybrid genetic programming algorithm, to obtain a corresponding target assembly sequence.

7. An electronic device, comprising: The application relates to a method for solving a complex product assembly sequence planning problem considering human factors, comprising: A memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for solving a complex product assembly sequence planning problem considering human factors according to any one of claims 1-5.

8. A computer readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the method for solving a complex product assembly sequence planning problem considering human factors according to any one of claims 1-5.

Images