Research On Multi-level Quality Control Modes And Control Charts Methods For Manufacturing System

The product conformity quality is the result of comprehensive effect of all sorts of errors or defects in manufacturing system.Especially,in terms of complex manufacturing system,the product quality is affected by many factors,whose relationships are complex.The characters of quality problems include diversity,hierarchy,randomness and unexpectedness.In many enterprises,quality problems occur frequently,and especially the repetitive quality events are common.Enterprises are struggling to cope with them.These have brought huge economic losses,and product quality cannot be effectively guaranteed and improved.So there is an urgent need to solve the problem for product quality control in manufacturing enterprises.This thesis is about typical quality problems in manufacturing system.The methods GQMM,fuzzy theory,evidence theory,Markov chain are induced for systematic and structured modeling of the quality problem,the design and application of control charts.Full eight chapters:In Chapter 1,scientific backgrounds and the state-of-the-art about quality control technology for manufacturing system are comprehensively elaborated.And then the bearing manufacturing process is introduced.The characteristics of quality control technology for manufacturing system are also analyzed.The structure and contents of this thesis are depicted at the end of this chapter.In Chapter 2,the structured analysis method for quality problems is proposed.It is a multi-level and knowledge-based quality control model which takes Goal,Question,Metric,Measure and other fundamental elements as the main line,structured principle as guidance,manufacturing process as carrier,and improving quality of manufacturing system as destination.Firstly,a structured analysis principle GQMM is introduced.Secondly,a structured analysis method for quality problems is proposed.Thirdly,a formal description of mappings between GQMM is also discussed.All these lay the foundation for the quality control mode decision and statistical process control technology development.Finally,the method is applied to bearing manufacturing process.Chapter 3 is for enterprises continuously falling into fire-fighting management because of quality problems' emerging endlessly.To improve the situation,this thesis proposes multi-level quality control modes which include four typical quality control modes:reactive,preventive,predictive and proactive.Firstly,the characteristics of quality problems are summarized as:hierarchically structured,causal,dynamic.Secondly,to control product quality effectively,the multi-level quality control modes are presented.And the characteristics of multi-level quality control modes are analyzed and comprised.Finally,the process and the frame of the multi-level quality control modes are discussed.Chapter 4 proposes a quality control decision method based on fuzzy cognitive map and evidence theory.Firstly,the four typical quality control modes and decision factors are analyzed;then,the quality control decision model and implementation steps are presented.Aiming at solving quality problems scientifically,the decision network is constructed and the most appropriate quality control mode is chosen for actual requirements.In order to construct multi-expert fuzzy cognitive map(FCM),we use multi-expert's knowledge as evidences,the possible value of weight as frame of discernment,and use evidence rule combing to give fusion basic probability assignment.Finally,the proposed method is applied to the bearing groove shape error out of specification limits problems in the bearing enterprises,which proved that the method was effective.In Chapter 5,the fuzzy univariate control charts are put forward to solve the problem that conventional control charts can't be applied to control fuzzy quality characteristics.Firstly,fuzzy quality characteristic are translated to representative statistics which are fuzzy mode transformation,fuzzy?Level midrange transformation and fuzzy a Level median transformation.And control charts are designed based on Poisson distribution.Secondly,the effects of the different statistics are analyzed.Direct fuzzy control chart is designed to avoid some information omission when translating fuzzy quality characteristic into representative statistics.The area ratio that fall within the control limits is used to conclude whether the process is relative out of control or in control.The performance of the control charts is analyzed by Matlab simulation.Finally,the example about watt-hour meter assembly is given to prove the proposed method.Chapter 6 is for quality characteristics which are difficult to measure and describing in fuzzy multi levels.To solve the problem of lower power when implementing statistical process control to these quality characteristics,F-MEWMA is put forward.F-MEWMA tries to make full use of multi levels process quality information based on fuzzy theory.Since different?cut sets containing different amount of information,interval value of weighted fuzzy?cut set is proposed.Mathematical characterization and comparison of the quality characteristics for fuzzy univariate data and fuzzy multivariate data are carried out respectively,and the control limits for different weight coefficients ? and different dimensions P are determined based on Markov Chain method.Then,F-EWMA and F-MEWMA are designed.The effect of F-MEWMA is analyzed by calculating probability of identifying variation by Matlab simulation.Finally,the implementation of the F-MEWMA on the electric energy meter running and starting quality characteristics showed good result.Chapter 7 focuses on the application of reactive,preventive and predictive quality control technology.The application of reactive quality control technology for manufacturing miniature bearing ferrule shows how to solve the quality problem quickly.The application of preventive quality quality monitoring technology shows how to apply SPC for geometry dimension,surface roughness and roundness.And the application of predictive quality monitoring technology is monitoring four parameters including cross feed,spindle motor power,grinding wheel vibration and acoustic emission.Finally the relationship is built between the features of the four parameters and the quality of groove grinding.And the relationship provides the foundation for concluding bearing grinding process level and adjusting the equipment and process parameters to enhance the quality monitoring of groove grinding.Chapter 8 gives the conclusions of this dissertation research,prospect for further research in this field.