Analysis Of Manufacturing System Complexity And Influencing Factors

With the improvement of living standards, customers' demand for the products are changing toward diversification and novelty. Traditional mode of automatic line suitable for mass production can not meet the requirements of enterprises. Enterprises must find new production technology to meet the market demand of diversity and small and medium-batch size. In such cases, advanced manufacturing technology came into being such as flexible manufacturing, lean manufacturing, ect.Advanced manufacturing technology generation and application enhances the dynamic characteristics of the manufacturing environment, which makes manufacturing systems become more complex and increases the frequency of decision-making in production planning and control decision-making. Meanwhile, high system integration induced by the application of advanced manufacturing technology makes managers difficult to predict its impact on system performance. Study found that many systems' flexibility are subject to the complexity of decision-making and control, in other words, flexible system is not flexible if before making a decision managers can not correctly comprehend and control the system complexity.So, effective measurement and influencing factors study not only be convenient to capture control point for managers and provides basis and method to compare different system complexity for planners.Currently information entropy method that researches complexity in terms of weighing system's uncertainty is used to measure and compare complexity of manufacturing systems, which has obtained extensive attention and common consensus, but some limitations arise in practical application. Meanwhile, previous studies have focused on a particular element of the complexity and all the major factors and its impact were not taken into account.Based on the above problems, the paper first summarized the studies of four areas at domestic and abroad including manufacturing systems complexity definition, influencing factors, measurement methods and relationship with system performance. For the current limitations of study views and methods, the main influencing factors are identified and measurement model is developed which effectiveness of the construction is validated by a computation example. On this basis, eight major factors in the model as the simulation variables, flow time deviation as a performance evaluation index, 8 simulation programs are determined by setting the variables' different level of the value. And EM-PLANT simulation software is applied to obtain system performance evaluation sample value under a different level of complexity factors. Finally, the analysis of system performance variance under different simulation programs is performed using analysis of variance(ANOVA) to test influencing level of various factors on system performance and determine sensitive factors.The results of simulation and ANOVA showed that: all of the eight factors in the mathematics model have impacts on system performance, thereby affecting the complexity. Increase of product mix, processing steps, the number of processing center, product structure breadth and the width can result in increased complexity. When product mix ration are equal, the impact on complexity is greater. When parts commonality and processing route commonality increase, system complexity decreases. As for signal factors analysis, product structure breadth and width is the major influencing factors which influence is even greater.Dissertation includes five chapters and key problems to be solved as followed: (1) identify the influencing factors of manufacturing system static complexity;(2) considering the impact of previously identified factors, develop simple and effective measure method of the manufacturing systems static complexity;(3) test the relationships between influencing factors and system performance and identify the sensitive factors to provide breakthrough for managers to improve system performance. The specific contents are as follows:Chapter 1, the introduction. The background, purpose, significance and contents of this research are introduced.Chapter 2, current research status of manufacturing system complexity in domestic and foreign countries. Four aspects from the definition of complexity, influencing factors, measurement methods and relationship with system performance in domestic and foreign research are summarized to examine the current advantages and disadvantages of study perspectives and methods about static complexity, which provide a theoretical basis for identifying influencing factors and establishing simple and effective measurement method.Chapter 3, method development of static complexity measurement. Based on limitations of study perspectives and methods in production application, the main influencing factors are determined and measurement model is developed which effectiveness of the construction is validated by a computation example.Chapter 4, influencing relationships analysis between complexity influencing factors and system performance. Eight major factors in the model as the simulation variables, flow time deviation as a performance evaluation index, 8 simulation programs are determined by setting the variables' different level of the value. And EM-PLANT simulation software is applied to obtain system performance evaluation sample value under a different level of complexity factors. Finally, the analysis of system performance variance under different simulation programs is performed using analysis of variance(ANOVA) to test influencing level of various factors on system performance and determine sensitive factors.Chapter 5, conclusion and prospect. which summarizes the main research and also sum up the novelties and problems requiring further research.Innovation of this dissertation is as follows:1, limitations in production application is improved, simple and practical static complexity measurement model is developed. The model has metrics of accessible computation data and easy to comprehend which provide standard for managers to evaluate and compare different system complexity.2, The main influencing factors previous references mentioned are included in the model, which reflect the actual situation of manufacturing systems and is convenient to trace the control points in complexity management.3, Simulation method is applied to explore the influencing relationships between complexity factors and system performance, which bridges the gap that an empirical study can not achieve effective control of variables change and exclude interference of dynamic environmental factors. So the new model is general applicability.