Cell formation and layout design for the blade manufacturing facility at DSME Trenton inc. or DSTN.

点击翻译按钮获取中文摘要

This is a proposal to extend the existing approaches in the literature for the design of discrete parts manufacturing systems. The industrial partner for this proposal is DSME Trenton or DSTN, a manufacturer of wind towers and turbines in Trenton, Nova Scotia. The work outlined in this proposal will be tested in their manufacturing facility for wind turbine production. Traditionally, high volume products with low variety are manufactured using product or flow lines. The job shop or process layout organization is used when product volumes are low and the variety high. For intermediate volume and variety, group technology or cellular manufacturing is the predominant design choice in industry. Other layout organizations considered relevant for intermediate volume and variety manufacturing are fractal layouts, holonic layouts, virtual manufacturing cells, etc. The main challenge in designing a manufacturing system are its stochastic and dynamic aspects. Stochastic aspects are important to take into account because there is uncertainty in product demand, product mix, product routings, availability of machines due to breakdown or preventive maintenance, and choice of manufacturing technologies available. The dynamic aspect refers to the change in the design of the facility over time. In our methodology, we propose two families of mixed-integer programming models for the design of a cellular manufacturing system. The first mixed-integer model is non-spatial and tries to solve the dynamic group technology cell formation problem where the objective is to minimize the sum of inter-cell transportation costs, machine acquisition costs, and the cost of relocating machines from one cell to another. The second mixed integer programming model is spatial and tries to come up with an optimal layout for a manufacturing system in each period. Both models will be built in the context of robust optimization, whose goal is to ensure that the designed system is stochastically robust. The models developed in this research will be applied at DSTN. In turn, the company will be the beneficiary of layout design recommendations for their wind turbine production facility.