A new statistical method for design and analyses of component tolerance

Subject Areas : Mathematical Optimization

Mohammad Mehdi Movahedi ¹ , Mohsen Khounsiavash ² , Mahmood Otadi ³ , Maryam Mosleh ⁴

1 - Department of Management, Firoozkooh Branch, Islamic Azad University, Firoozkooh, Iran
2 - Department of Electrical, Biomedical and Mechatronics Engineering, Qazvin Branch, Islamic Azad University, Qazvin, Iran
3 - Department of Mathematics, Firoozkooh Branch, Islamic Azad University, Firoozkooh, Iran
4 - Department of Mathematics, Firoozkooh Branch, Islamic Azad University, Firoozkooh, Iran

Received: 2020-10-04 Accepted : 2020-10-04 Published : 2017-03-01

Keywords: Estimates, Nonlinear programming  Generalized lambda, distribution (GLD)  Tolerancing  Percentile matching,

Abstract :

Tolerancing conducted by design engineers to meet customers’ needs is a prerequisite for producing high-quality products. Engineers use handbooks to conduct tolerancing. While use of statistical methods for tolerancing is not something new, engineers often use known distributions, including the normal distribution. Yet, if the statistical distribution of the given variable is unknown, a new statistical method will be employed to design tolerance. In this paper, we use generalized lambda distribution for design and analyses component tolerance. We use percentile method (PM) to estimate the distribution parameters. The findings indicated that, when the distribution of the component data is unknown, the proposed method can be used to expedite the design of component tolerance. Moreover, in the case of assembled sets, more extensive tolerance for each component with the same target performance can be utilized.