Qualitative Simulation Based Auto-Body Conceptual Assembly Process Planning And Decision-Making

Assembly variations would directly affect vehicle appearances and functions, so more and more experts and researchers study variation types and variation accumulation in each stage of vehicle design in order to make variation analysis and control. Due to incomplete information and undetermined and/or underdetermined parameters with a big design space and a large scope of solution exploration in the conceptual design,general numerical calculation methods (such as CATS, VSA and 3DCS) could not perform conceptual assembly variation simulation. Quantitative thinking methods of problem solvers and quantitative analytical tools can not solve uncertainty and incompleteness in the conceptual phase, it is necessary to employ limited information to reason so as to produce enough feasible alternatives, and make trade-off comparison to improve rationality and robustness of design concepts. The essential sources of many problems and errors which happen in the production,customer use and maintenance phases are concealed early in the conceptual design stage, so The dissertation aims to explore the application of qualitative simulation in the auto-body conceptual assembly design, and research qualitative tolerance analysis and qualitative multiple attribute decision-making. The primary studies of the reseach conducted here are following:First, the paper introduces multi-attribute di-graph and qualitative knowledge to accomplish auto-body conceptual assembly process modeling. After the accomplishment of qualitative knowledge from physical parameters of assembly features, the system should make formalized description of incomplete information system by means of semantics, symbolization and fuzziness, and get key influential factors of assembly process by qualitative analysis and generate the reasonable results through qualitative model by qualitative reasoning. Auto-body assembly modeling makes relation modeling, and builds relation between key characteristics, and then performs logic expression between design factors and objective variables. Further,the paper study modular assembly design in the conceptual design step, and make assembly constraint modeling through constraint classification. In the auto-body assembly process, the sound joint configuration would reduce and/or eliminate assembly variation propagation between parts, because rational allocation and design of joint configuration enable to adjust assembly process locators to achieve slip joint absorb or compensate the variation accumulation to improve the assembly robustness, therefore, the paper study joint type design and configuration and describe a multi-station auto-body assembly propagation model.Secondly, the qualitative simulation mechanism is constructed in the paper, one main task of which is to build qualitative simulation functions of assembly dimension chains. The goal of assembly qualitative simulation system is to construct a set of qualitative differential equations,and qualitative simulation is carried according to these qualitative equations, which are abstracted from quantitative tolerance functions, in order to predict assembly potential behaviors. Due to the incompleteness of information during early phase, the assembly process index is introduced to determine the base part of assembly to prune the assembly sequences. With the given assembly sequences and the tolerance generation of certain nominated key characteristics, the system should make obtainment and/or qualitative description of corresponding part parameters and feature attributes, and model qualitative mapping relation between functional characteristics and dimension constraints and geometrical constraints. These qualitative mapping relations are certain explicit or implicit mathematical functions, the system use these functions to perform variation analysis. The assembly qualitative analysis system builds mathematical logic models and/or qualitative models to do computer experiments to simulate the transition from an assembly state to another assembly state according to certain criteria, and employs reasonable joint configuration to fulfill active variation compensation. Auto-body conceptual assembly qualitative simulation procedure is proposed to conduct hierarchical qualitative simulation to analyze and predict variation flow with limited parameters and fuzzy knowledge.Finally, a qualitative multi-attribute decision-making model is constructed in the paper,which employs multi-attribute qualitative criteria to make selection and evaluation of early design concepts in order to provide helpful direction and auxiliary decision for the next design step. An optimization model integrated with assembly sequence, joint configuration and tolerance allocation is proposed to calculate auto-body conceptual assembly variation accumulation. Decision-making of auto-body assembly design system is to determine the optimal ones out of numerical feasible solutions, and conceptual assembly decision-making mainly performs qualitative multi-attribute evaluation of assembly solutions. With the help of designer interactive input or system heuristic generation of weight values of each attribute according to the corresponding importance, the qualitative decision method makes priority ordering to get the best assembly design solutions out of the solution set.On the basis of above researches, the auto-body conceptual assembly variation analysis system is constructeded. The auto-body assembly resultant variation is not only related with part manufacted quality and joint types between parts, but also is affected by assembly sequence, so the whole auto-body conceptual assembly process should gradually reasonably adjust and optimally combine the three design factors of assembly sequence, joint configuration and tolerance allocation to meet assembly key characteristic requirements and product final assembly quality. An object-oriented qualitative assembly tolerance design framework is proposed to provide a concurrent timely interactive environment to handle the incomplete information for designer.In a summary, a multi-attribute digraph conceptual assembly model is proposed and the qualitative simulation method is introduced to auto-body assembly design to study conceptual assembly variation accumulation and propagation. The obtainment and expression of assembly qualitative equations, qualitative reasoning mechanism and qualitative control strategies, all these are needed to construct the knowledge-based assembly design system prototype. To deal with those design characters of conceptual design phase, qualitative simulation based auto-body assembly variation analysis and decision-making methodology is proposed, which transfers knowledge of designers and practical experience of factories to qualitative knowledge, and which uses knowledge reasoning to make assembly problem-solving and decision optimization, naturally, a knowledge-based conceptual assembly design method is introduced to make assembly quality control and prediction in the early design stage so as to find feasible assembly concepts, and to increase robustness of assembly design solutions and to reduce cycle of product development.