Statistical Adjustment and Calibration of Complex Systems considering Multiple Outputs: Case Study of Laser-Assisted Micromachining Process

Abstract

By rapid advancements in technologies, studying and simulating a real complex system taint with uncertain parameters is too demanding. Based on the relevant literature, there are three approaches to recognize and simulate different systems: engineering, statistical, and engineering-statistical approaches. Regarding the purpose of this research, the engineeringstatistical approach is applied. By considering the single output, Laser Assisted Micro-Machining (LAMM) is studied using the engineering-statistical approach. In the present study, * Corresponding Author; Vanak Sq., pardis, St., K. N. Toosi University; 2 an engineering-statistical approach is applied to LAMM, studying two outputs simultaneously. To investigate variates simultaneously, there are some complicated issues, such as calibrating variates at the same time, adjusting them concurrently, and calculating the values of parameters, with which this paper should cope. Considering the Mean Squared Prediction Error (MSPE) as the comparison index, for thrust force output, the index value was obtained as 1.48 by Kennedy and O'Hagan model, 2.47 by the model presented by Roshan and Yan, and 1.9425×10 -4 by the proposed model. Moreover, for cutting force output, the index was obtained as 0.21 by the Kennedy and O'Hagan, 1.41 by Roshan and Yan, and 1.6×10 -8 by the presented model. The obtained values demonstrate reasonable and comparable results for the MSPE index, compared with the models considering the outputs individually.