An interactive on-line multi-objective optimization approach with application to metal cutting turning operation

Abstract

An on-line supervisory computer control model for optimization of multiple objective programming problems is developed and applied to machining problems. Specifically, the turning operation is formulated and solved by the approach. For the first time, the characteristics of on-line multi-objective problems are described and an approach for solving such problems is developed. The proposed system maintains feasibility of operations and optimizes objectives based on the machine performance and on-line interactions with the decision-maker (DM). The model is hierarchical and consists of two levets: level 1 is designed to protect automated machinery and the workpiece by monitoring the process outputs; and level 2 is designed to optimize the on-line multi-objective problem. The tasks of both levels are accomplished by on-line monitoring and control of changes in process outputs via real-time information provided by sensors. The recommended parameters, obtainable from the multi-objective optimization approach, are used to create and update a database for machinability problems for future use. An example is discussed.