Product Family Planning Method Based On Multi-objective Genetic Algorithm

With the continuous expansion of product line of manufacturing enterprise and constantly differentiation of product type,many phenomenon of diminishing marginal returns appear.Growth of product sales and additional revenue is not synchronized with the increase of product variety,product family design become a problem that can not be ignored.Designing and developing product family is recognized to be a kind of effective methods to achieve economies of scale,it is used to adapt to increasing product varieties in different market niche.In addition to reducing costs in providing diversified products through using proven elements in a company's activities and provided resources,product family design can also reduce development risk and system complexity,improve the ability to upgrade the product,improve the flexibility and responsiveness of the manufacturing process.This dissertation impact electric drill,hydraulic excavator and general motors as the research object.After summarizing research progress of multi-objective genetic algorithm and product family planning,the multi-objective genetic algorithm was applied to typical modular and parametric products to ensure a better product development efficiency.Factors that affect module planning were classified aiming at modular product,three principles of personalized parts separation principles,variable component separation principles and largest degree of internal polymerization and minimum degree of connection between modules were proposed.Different rules were coded in fitness function of multi-objective genetic algorithm through Matlab simulation,different rules domination results were obtained through module partition.Advantages of this approach were expounded combined with the application of impact electric drill.The internal and external driving factors of product formation were summarized on basis of function module division.Three targets of machine performance,cost and delivery time were proposed,three targets were used as fitness function of multi-objective genetic algorithm through Matlab simulation,module configuration schemes of different dominant principles were obtained.This approach was expounded combined with hydraulic excavator.General optimization model of parametric product family was established aiming at topology of the parametric product.Preliminary design parameters are acquired by using multi-objective genetic algorithm through Matlab simulation.Multivalued platform manual operation method of design parameters were explored.The effectiveness of the method was demonstrated using general motors product family as an example,platform index was established to verify its advantage.