Reliability modeling-based tolerance design and process parameter analysis considering performance degradation
Introduction
Tolerance design and process parameter analysis have been extensively studied in the literature, and they are two effective methods to reduce fluctuations in product quality characteristics (QCs). Tolerance design is to determine the allowable fluctuation range of QCs, so as to minimize the manufacturing cost and ensure the normal function of products [1], while process parameter analysis is performed to improve the manufacturing process so that QCs of products can be within the given tolerance [2]. Most existing works focus on finding the best compromise between manufacturing cost and tolerance of QCs, regardless of the degradation effect. For some products, their QCs will continue to degrade during the use period, result in that the QCs are unable to remain within the initial tolerances and ultimately affect the normal operation of products. Therefore, ignoring the degradation effect will lead to that the tolerance design and process parameter analysis cannot meet the high-reliability requirement. For example, the free length is the most important QC of helical springs that are commonly used in mechanical devices such as spring-loaded safety valves (SLSVs) [3], and the free length determines the spring force provided by the helical spring when it is compressed to a fixed height. Due to the stress relaxation, the free length of helical springs will continue to degrade during the use period until exceeding the design specification, which will fail the mechanical device. Motivated by this illustrative example, this paper intends to develop a more reliable tolerance design and process parameter analysis considering the degradation performance of product QCs.
Tolerance of products refers to the allowable fluctuation range of QCs in terms of economy and function. Generally, limiting the fluctuation range implies the improvement in product quality and the increase in manufacturing cost [4]. To date, the studies on the tolerance design of products have been widely discussed in the literature, Hallmann et al. [5] provided a review of the state of the arts in the tolerance design, where the manufacturing cost and quality cost are commonly considered as the main constraints. Recent developments in tolerance design heightened the necessity of considering reliability constraints. Wang et al. [6] pointed out the importance of reliability in the tolerance design of DC hybrid contactor, and they developed a reliability tolerance design based on EDA software. From then on, the tolerance design considering reliability constraint has been applied to other products and manufacturing processes, such as mechanical assemblies [7], switched-mode power supply [8], and CNC machine tools [9]. However, all the literature mentioned above only focused on the influence of tolerance on the design reliability, i.e., the ability of the tolerance of QCs to meet the design requirement. For many products like helical springs, as the use time increases, their QCs will degrade during the use period, and result in the performance deviating from design specifications. Therefore, a more significant constraint on the tolerance design is the operational reliability, which is defined as the ability of products to work without failure under the stated conditions for a specified period. In view of this, Zhai et al. [10] developed a tolerance design of electronic circuits considering performance degradation and carried out an optimum allocation based on the predicted performance under the given task condition and task time. Nevertheless, the effect of the initial value of QCs on the degradation process is ignored in their works. For helical springs, when they are compressed into a fixed height, different initial free lengths will cause different mechanical stresses on helical springs, and the corresponding degradation performance will also show differences. Confronted with that, based on the initial tolerance designed from functional and economic perspectives, and considering the operational reliability constraint, this paper intends to develop a more reliable tolerance design of products based on the performance degradation affected by the initial value of QCs.
In addition, process parameters refer to manufacturing process control parameters that have a significant impact on the product quality, such as the axial pressure, initial rotational speed, and moment of inertia of the flywheel in solid-state welding processes [11]. Process parameter analysis aims to optimize manufacturing process control parameters to minimize the variance in QCs and the difference between the real value and the target value of QCs (i.e., the deviation in QCs) [2]. Recently, many different optimization models have been developed for this goal in different fields, such as the energy model developed by Bi and Wang [12] for the optimization of energy consumption in the production process, the SS-SVR meta-model studied by Zhang and Zhou [13] for the parameter optimization of laser magnetic welding process, and the gray relational analysis used in Yu et al. [14] for the process parameter optimization during the polishing process of optical components. However, to our best knowledge, existing works mainly focused on minimizing the deviation or variance of QCs and rarely considered how small the deviation and variance of QCs should be so that the manufacturing process can be accepted (i.e., the products produced by the manufacturing process satisfy the given requirement). In view of this, considering the relationship between QCs and the operational reliability of products, this paper aims to derive an allowable fluctuation range for the deviation and variance of QCs from the perspective of operational reliability.
Reliability modeling is the core tool to evaluate the operational reliability of products, and relevant studies have been widely developed in recent years. Some recent reviews of degradation modeling are available at Ye and Xie [15], Zhang et al. [16], and Kang et al. [17]. In general, there are three typical methods for degradation modeling: degradation amount distribution model [18], stochastic model (including Wiener process [19], Gamma process [20], and inverse Gaussian process [21]), and general path model. Among those models, the general path model, developed by Lu and Meeker [22] firstly, has become the most widely used degradation model as it is flexible to model the unit-to-unit variation and measurement errors. Recent developments in general path model include the multivariate path model proposed by Si et al. [23] and Lu et al. [24] for the reliability analysis of systems with multiple characteristics, and the linear path model embedded with self-recovery mechanism considered by Liu et al. [25] and Kong and Yang [26]. Therefore, considering the high applicability of the general path model, and inspired by Yuan et al. [27], this paper uses the general path model to describe the degradation performance of products. Furthermore, for highly reliable products like helical springs, it is still a challenge to obtain sufficient degradation data in a short time. Then, the accelerated degradation test (ADT) was developed to solve this problem. ADT is carried out by exposing and testing the product under high stressed conditions. Limon et al. [28] provided a review of the state of the arts in the optimal design of ADT, where the constant stress accelerated degradation test (CSADT) is considered the most primary and effective method [29,30]. The key to CSADT is to find the accelerated stresses that are sensitive to the product degradation. As the primary failure mode of helical springs is stress relaxation, many studies have shown that temperature and humidity have a significant effect on the stress relaxation, see Luo et al. [31] and Li et al. [32] for examples. Therefore, in this paper, temperature and humidity are listed as the principal accelerated stresses.
Consequently, motivated by the stress relaxation process of helical springs, this paper intends to develop a more reliable tolerance design of products considering performance degradation, and further derives an allowable fluctuation range for the deviation and variance of QCs to guide the process parameter analysis and improve the manufacturing process of products. In addition, considering the difficulty in obtaining the degradation signals, an accelerated degradation model is proposed by embodying the temperature-humidity-mechanical stress covariate into the general path model, through which the influence of the initial value of QCs on the operational reliability of products can be extracted from the CSADT.
The major contributions of this work are summarized as follows:
- (1)
In order to make the tolerance of products meet the high-reliability requirement, the proposed tolerance design takes into accordant with the performance degradation of products and the influence of the initial value of QCs.
- (2)
The proposed process parameter analysis quantifies the influence of process fluctuation on the operational reliability of products, and derives an allowable fluctuation range of the process fluctuation, which is critical for the improvement of the manufacturing process.
- (3)
While giving the tolerance design and process parameter analysis of helical springs, the proposed method also provides a framework of the tolerance design and process parameter analysis for products with degradation performance.
The rest of this paper is organized as follows. The detailed construction of the accelerated degradation model with the effect of initial free length and the estimation method of unknown parameters are presented in Section 2. Then, based on the accelerated degradation model, the tolerance design and process parameter analysis considering the reliability constraint are introduced in Section 3. Moreover, In Section 4, a real CSADT is designed to generate degradation information of helical springs, through which the implementation and effectiveness of the proposed method are illustrated. Finally, some conclusions are given in Section 5.
Section snippets
Accelerated degradation model with the effect of initial free length
For the helical spring with a fixed initial free length , when it is compressed to a specified height h, the initial deformation and the corresponding mechanical pressure are constant. This section first presents a new accelerated degradation model for helical springs considering the environmental stress and measurement errors, along with the reliability assessment results with temperature-humidity-mechanical stress covariate. Then, based on nonlinear least square (NLS) and maximum
Tolerance design and process parameter analysis considering reliability constraint
Based on the accelerated degradation model developed in Section 2, the tolerance design of helical springs considering reliability constraint is first carried out in this section. Then, for the process parameter analysis, since it mainly focuses on the improvement of the manufacturing process, the fluctuation in the initial free length is used to characterize the quality of the manufacturing process, and a new reliability model is proposed to depict the influence of parameter fluctuation on the
Case study
In this section, a real CSADT is firstly designed to generate the degradation information of helical springs. Then, the tolerance design based on performance degradation developed in Subsection 3.1 is used to carry out the tolerance design of the initial free length. Finally, the process parameter analysis proposed in Subsection 3.2 is used to derive the allowable fluctuation range for the deviation and variance of the initial free length, which can guide the improvement of the manufacturing
Conclusion
Motivated by the stress relaxation process of helical springs, and focusing on the performance degradation and the influence of the initial value of QCs, this paper developed a tolerance design and process parameter analysis of products subject to the operational reliability requirement. Given the difficulty in obtaining degradation information of high-reliability products like helical springs, a reliability model considering the accelerated performance degradation and the influence of the
CRediT authorship contribution statement
Xuefeng Kong: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing - original draft, Writing - review & editing, Visualization, Project administration. Jun Yang: Conceptualization, Resources, Writing - review & editing, Supervision, Project administration, Funding acquisition. Songhua Hao: Conceptualization, Methodology, Writing - review & editing, Project administration.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (46)
An analytical study of the dynamics and stability of a spring loaded safety valve
Nucl Eng Des
(1982)- et al.
A geometric accuracy analysis and tolerance robust design approach for a vertical machining center based on the reliability theory
Measurement
(2020) - et al.
Study on optimization of ultrasonic-vibration-assisted polishing process parameters
Measurement
(2019) - et al.
Model-driven degradation modeling approaches: Investigation and review
Chin J Aeronaut
(2020) - et al.
Reliability modeling for a two-phase degradation system with a change point based on a Wiener process
Reliab Eng Syst Saf
(2020) - et al.
Reliability analysis of systems considering clusters of dependent degrading components
Reliab Eng Syst Saf
(2020) - et al.
Degradation analysis based on an extended inverse Gaussian process model with skew-normal random effects and measurement errors
Reliab Eng Syst Saf
(2019) - et al.
Reliability modeling for dependent competing failure processes of damage self-healing systems
Comput Ind Eng
(2017) - et al.
Reliability analysis of composite insulators subject to multiple dependent competing failure processes with shock duration and shock damage self-recovery
Reliab Eng Syst Saf
(2020) - et al.
Optimal plan for Wiener constant-stress accelerated degradation model
Appl Math Model
(2020)
Investigation on high-temperature stress relaxation behavior of Ti-6Al-4V sheet
Mater Sci Eng A
Microstructure evolution in stress relaxation behavior of austenite AISI 304 stainless steel spring
Mater Charact
Stress relief effect on fatigue and relaxation of compression springs
Mater Des
A Handbook for Developing Optimal Specifications
Technometrics
Modern Industrial Statistics: Design and Control of Quality and Reliability
Technometrics
Tolerances analysis and cost evaluation for product life cycle
Int J Prod Res
From tolerance allocation to tolerance-cost optimization: a comprehensive literature review
Int J Adv Manuf Technol
Method of reliability tolerance design based on EDA technology and its application on DC hybrid contactor
J Zhejiang Univ-Sci A
Tolerance–reliability analysis of mechanical assemblies for quality control based on Bayesian modeling
Assem Autom
Tolerance-based reliability and optimization design of switched-mode power supply
Qual Reliab Eng Int
A Tolerance Design Method for Electronic Circuits Based on Performance Degradation
Qual Reliab Eng Int
Process parameter analysis of inertia friction welding nickel-based superalloy
Int J Adv Manuf Technol
Optimization of machining processes from the perspective of energy consumption: A case study
J Manuf Syst
Cited by (14)
An Optimal Tolerance Design Approach of Robot Manipulators for Positioning Accuracy Reliability
2023, Reliability Engineering and System SafetyProcess parameter optimization for lifetime improvement experiments considering warranty and customer satisfaction
2022, Reliability Engineering and System SafetyCitation Excerpt :In the manufacturing phase, many processes may affect the final lifetime of product significantly [1, 2].
Optimal design of step-stress accelerated degradation test oriented by nonlinear and distributed degradation process
2022, Reliability Engineering and System SafetyCitation Excerpt :Similar to the above method, the virtual output characteristics of the LED driver were obtained based on the degradation distributions of the sensitivity components parameters. Kong et al. [29] implemented the reliability modeling-based tolerance design for a helical spring, where the reliability model was deduced based on the degradation distribution parameters of the stress relaxation of the helical spring. The above works conclude that the degradation distribution of products is an important issue.
A Bayesian-reliability based multi-objective optimization for tolerance design of mechanical assemblies
2021, Reliability Engineering and System SafetyCitation Excerpt :Anwer et al. [20] investigated the fundamentals of the skin model at a conceptual, geometric, and computational level as well. Kong et al. (2020) proposed an approach for the tolerance design and process parameter analysis that can consider performance degradation [21]. Moreover, in this work, a reliability model with temperature-humidity-mechanical stress covariate has been presented.
Analysis and Optimization of Adjust-Free Rate of Balanced Force Relay Contact System
2024, Lecture Notes in Electrical EngineeringNon-traditional tolerance design techniques for low machining cost
2023, International Journal on Interactive Design and Manufacturing