1 Department of Manufacturing Engineering, Technical University of Denmark2 Department of Management Engineering, Technical University of Denmark
This thesis deals with the adaptation and implementation of various optimisation methods, in the field of experimental design, for the laser cutting process. The problem in optimising the laser cutting process has been defined and a structure for at Decision Support System (DSS) for the optimisation of the laser cutting process has been suggested. The DSS consists of a database with the currently used and old parameter settings. Also one of the optimisation methods has been implemented in the DSS in order to facilitate the optimisation procedure for the laser operator. The Simplex Method has been adapted in two versions. A qualitative one, that by comparing the laser cut items optimise the process and a quantitative one that uses a weighted quality response in order to achieve a satisfactory quality and after that maximises the cutting speed thus increasing the productivity of the process. Both versions have been used for optimisations in different materials and both of them succeeded in obtaining a satisfactory quality. In addition the optimisations with the quantitative version also obtained a much higher productivity . The results of the optimisations showed that the qualitative version mainly should be used when either it is not possible to measure the quality characteristics or if a quantative quality response has not yet been created and tested.A second metod was adapted for the laser cutting process and implemented in the DSS. This was the EVolutionary OPeration (EVOP) method which optimises the productivity without interrupting the production. The method was tested in three Danish entreprises using laser cutting in production. Unfortunately none of the optimisations were finished, since it turned out to be difficult to convince the laser operators to use the method regularly. This is mainly because a parameter setting that results in a satisfactory quality already exists when the optimisation is started.