Study On The Similar Task Assignment Method Based On Learning Effect

With the continuous expansion of market competition among manufacturers,under the premise of ensuring high-quality,high-efficiency,and low-cost production of products,how to use advanced scientific and technological means to achieve a reasonable task distribution is a key issue for companies and many scholars.This paper constructs a predicting model that can estimate task execution time and resolved effectively the problem that task execution time is a fixed value by combining with similarity theory and learning effect model.Combine the predictive model with intelligent optimization algorithms to assign product process design tasks.Finally,the prototype system of task executive module was constructed according to companies' actual demands,and validated the effectiveness and feasibility of the algorithm through examples.The main work of this research is as follows:(1)According to the characteristics of product process design and take the similarity theory between process design tasks as the research object,compared and analyzed existing similarity calculation theoretical model and applicable objects.Based on the product process design task,built a similarity calculation model,This model breaks down the similarity calculation of the process design task into calculations of document similarity and model similarity.The influence of similarities in process design tasks on the time of task execution be was demonstrated by instances.(2)Combining the similar task assignment method of product process design,considering the individual ability of process design executives,analyzing the existing learning effect models and application fields,expounded the influence of the process designer's learning effect on task execution time,and constructed a predictive model that can estimate the execution time of similar tasks in process design.This model combines the S-Model with the similarity calculation model,and considers the proportion of the computer-aided in executions.Finally,the validity of the model and algorithm was is tested by the comparative analysis experiment of enterprise's actual case and regression analysis of MATALB.(3)According to the existing research on the similar task assignment of process design,built the combination model through the adaptive genetic algorithm and the similar task execution time prediction model,and formed a set of similar task assignment algorithm based on learning effect.The algorithm uses an improved circular crossover operator to optimize the performance of the improved genetic algorithm.Finally,the process design task assignment of an aerospace manufacturing company was used as the carrier to design the comparative analysis experiment.The two groups of comparative experiments were verified by MATLAB to verify the efficiency of the improved allocation algorithm,and the improved task allocation results and efficiency were achieved excellent.(4)According to the production needs of an aerospace manufacturing enterprise,the similar task assignment algorithm based on the learning effect is effectively integrated with the computer aided technology.Using the prototype system development platform,a prototype system that integrates process design,task assignment and execution into one is constructed,and it was experimentally verified.