A multi-objective problem based on fuzzy inference with application to parametric design of an electrophotographic system

This paper proposes a systematic approach to conduct system-level optimal parametric design for a class of electrophotographic systems. A conventional monochrome laser printer serves as the platform for verification of the proposed optimal design approach. Besides performance, we incorporate two other practical indices, i.e., cost and energy consumption, into design objectives and formulate a multi-objective optimization problem. A fuzzy inference system is established to provide the nonlinear or linguistic mapping between the decision variables and the design objectives. For comparative purpose, the problem is solved using both single-objective and multi-objective optimization algorithms. Note that the proposed approach is also applicable to other complex systems.

Research highlights
► A systematic approach to conduct system-level optimal parametric design for a class of electrophotographic systems is proposed.
► A multi-objecitve optimization problem is formulated, which incorporates two other practical indices, i.e., cost and energy consumption, into design objectives besides performance index.
► The novelty lies in constructing a fuzzy inference system to establish the nonlinear or linguistic mapping between the decision variables and the design objectives.
► The optimization problem is solved numerically with comparison and discussion of the utilized search algorithms and obtained solutions.