Intelligent Decision Making in Quality Management

Intelligent techniques present optimum or suboptimal solutions to complex problems, which cannot be solved by the classical mathematical programming techniques. The aim of this chapter is to review the intelligent decision making literature in order to reveal their usage in quality problems. We first classify the intelligent techniques and then present graphical illustrations to show the status of these techniques in the solutions of quality problems. These graphs display the publishing frequencies of the intelligent quality management papers with respect to their countries, universities, journals, authors, types (whether it is a conference paper, book chapter, journal 1 paper, etc.)

Shewhart’s control charts are used when you have enough and exact observed data. In case of incomplete and vague data, they can be still used by the help of the fuzzy set theory. In this chapter, we develop the fuzzy control charts for variables, which are namely \( \overline{X} \) and R and \( \overline{X} \) and S charts. Triangular fuzzy numbers have been used in the development of these charts. Unnatural patterns have been examined under fuzziness. Besides, fuzzy EWMA charts have been also developed in this chapter. For each fuzzy case, we present a numerical example.

Control charts for attributes are used to detect nonrandom variation when the inspected quality characteristic cannot be represented numerically. Fuzzy attribute control charts allow flexibility in evaluating whether an item is conforming or nonconforming. Thus, it is preferred when there is ambiguity about the conformity of the item. In this chapter, crisp attribute control charts, fuzzy attribute control charts and some numerical examples are given.

In this chapter we consider the economical design of EWMA zone control charts for set of machines operating under JPS (Jidoka Production System). We provide an extensive literature review of intelligent systems in quality control deductively to fit our purposes. It starts with an overview of quality control charts; then, reviews charts designed for special purposes such as EWMA, CUSUM and zone control charts. Finally, as particularly related to this study, reviews of economical design and intelligent applications of EWMA are provided. We discuss and review Jidoka Production System and motivation of operating such a system. We suggest an intelligent control and repair system such that in a production system, machines are individually controlled and repaired when an out-of-control signal is triggered in the zone with the tight control limits, however a system-wide shut down and repair is conducted when the out-of-control signal is from beyond the inner (tight) control limits which is considered as an opportunity for repair and calibration of all machines. We illustrate and investigate the behaviour of control parameters, namely sample size, sampling interval and control limits, via a numerical study of a three-machine system through simulation. We also provide insights for implementation of several metaheuristics for the system setting discussed in this chapter.

After the fuzzy set theory was introduced and developed, many studies have been realized to combine quality control methods and fuzzy set theory. This chapter is including the categorization of most essential works on fuzzy process capability indices in the following four main categories: After presenting the basic idea of the main works, all related studies briefly reviewed in each class. Some numerical examples are presented to show the applicability of the proposed methods.

In the design of critical combinations and complex integrations of large engineering systems, their reliability, availability and maintainability (RAM) analysis of the inherent processes in the system and their related equipments are needed to be determined. Although there have been tremendous advances in the art and science of system evaluation, yet it is very difficult to assess these parameters with a very high accuracy or precision. Basically, this inaccuracy in assessment stems mainly from the inaccuracy of data, lack of exactness of the models and even from the limitations of the current methods themselves and hence management decisions are based on experience. Thus the objective of this chapter is to present a methodology for increasing the performance as well as productivity of the system by utilizing these uncertain data. For this an optimization problem is formulated by considering RAM parameters as an objective function. The conflicting nature between the objectives is resolved by defining their nonlinear fuzzy goals and then aggregate by using a product aggregator operator. The failure rates and repair times of all constituent components are obtained by solving the reformulated fuzzy optimization problem with evolutionary algorithms. In order to increase the performance of the system, the obtained data are used for analyzing their behavior pattern in terms of membership and non-membership functions using intuitionistic fuzzy set theory and weakest t-norm based arithmetic operations. A composite measure of RAM parameters named as the RAM-Index has been formulated for measuring the performance of the system and hence finding the critical component of the system based on its performance. Finally the computed results of the proposed approach have been compared with the existing approaches for supremacy the approach. The suggested framework has been illustrated with the help of a case.

In this chapter, we briefly introduce the topic of acceptance sampling. We also examine acceptance sampling plans with intelligent techniques for solving complex quality problems. Among intelligent techniques, we focus on the application of the fuzzy set theory in the acceptance sampling. Moreover, we propose multi-objective mathematical models for fuzzy single and fuzzy double acceptance sampling plans with illustrative examples. The study illustrates how an acceptance sampling plan should be designed under fuzzy environment.

Experimental design (DOE) is a well-developed methodology that has been frequently adopted for different purposes in a wide range of fields such as control theory, optimization, and intelligent decision making. The main objective of DOE is to best select experiments to estimate a set of parameters while consuming as little resources as possible. The enrichment of literature on computational intelligence has supported DOE to extend its sphere of influence in the past two decades. Specifically, the most significant progress has been observed in the area of optimal experimentation, which deals with the calculation of the best scheme of measurements so that the information provided by the data collected is maximized. Nevertheless, determining the design that captures the true relationship between the response and control variables is the most fundamental objective. When deciding whether a design is better (or worse) than another one, usually a criterion is utilized to make an objective distinction. There is a wide range of optimality criteria available in the literature that has been proposed to solve theoretical or practical problems stemming from the challenging nature of optimal experimentation. This study focuses on the most recent applications of DOE related to heuristic optimization, fuzzy approach, and artificial intelligence with a special emphasis on the optimal experimental design and optimality criteria.

The ISO 9001:2008 quality management standard states that organizations shall plan and implement monitoring, measurement, analysis and improvement processes to demonstrate conformity to product requirements. According to the standard, detailed analysis of data is required for this purpose. The analysis of data should also provide information related to characteristics and trends of processes and products, including opportunities for preventive action. The preliminary aim of this chapter is to show how intelligent techniques can be used to design data–driven tools that are able to support the organization to continuously improve the effectiveness of their production according to the Plan—Do—Check—Act (PDCA) methodology. The chapter focuses on the application of data mining and multivariate statistical tools for process monitoring and quality control. Classical multivariate tools such as PLS and PCA are presented along with their nonlinear variants. Special attention is given to software sensors used to estimate product quality. Practical application examples taken from chemical and oil and gas industries illustrate the applicability of the discussed techniques.

The multidimensional nature of risks as well as substantial uncertainties and subjectivities inherent in the risk assessment process led a growing number of researchers to develop alternative approaches for failure mode and effects analysis. The purpose of this chapter is to provide a comprehensive review of the multi-criteria approaches proposed for failure mode and effects analysis under uncertainty and offer a brief tutorial for those who are interested in these approaches.

Quality function deployment (QFD) is commonly used in the product planning stage to define the engineering characteristics and target value settings of new products. However, some QFD processes substantially involve human subjective judgment, thus adversely affecting the usefulness of QFD. In recent years, a few studies have been conducted to introduce various intelligent techniques into QFD to address the problems associated with subjective judgment. These studies contribute to the development of intelligent QFD. This chapter presents our recent research on introducing intelligent techniques into QFD with regard to four aspects, namely, determination of importance weights of customer requirements, modeling of functional relationships in QFD, determination of importance weights of engineering characteristics and target value setting of engineering characteristics. In our research, a fuzzy analytic hierarchy process with an extent analysis approach is proposed to determine the importance weights for customer requirements to capture the vagueness of human judgment and a chaos-based fuzzy regression approach is proposed to model the relationships between customer satisfaction and engineering characteristics by which fuzziness and nonlinearity of the modeling can be addressed. To determine importance weights of engineering characteristics, we propose a novel fuzzy group decision-making method to address two types of uncertainties which integrates a fuzzy weighted average method with a consensus ordinal ranking technique. Regarding the target value setting of engineering characteristics, an inexact genetic algorithm is proposed to generate a family of inexact optimal solutions instead of determining one set of exact optimal target values. Possible future research on the development of intelligent QFD is provided in the conclusion section.

Six Sigma is a well-regarded and proven methodology for improving the quality of products and services by removing inconsistencies in processes. Insomuch of the early Six Sigma initiatives was focused on process effectiveness in meeting quality expectations and process efficiency for achieving maximum producer value; the future trends is moving towards utilizing feedback loops to create intelligent processes that enhances the adaptability to changing conditions. The purpose of this chapter is to extend understanding of what performance measures can be applied to processes in order to gain useful information and the emerging application of artificial neural networks to handle concurrent multiple feedback loops.

The Taguchi method has been widely applied in quality management applications to identify and fix key factors contributing to the variations of product quality in manufacturing processes. This method combines engineering and statistical methods to achieve improvements in cost and quality by optimizing product designs and manufacturing processes. There are several advantages of the Taguchi method over other decision making methods in quality management. Being a well-defined and systematic approach, the Taguchi method is an effective tuning method that is amenable to practical implementations in many platforms. To build on this, there are also merits, in terms of overall system performance and ease of implementation, by utilizing the Taguchi method with some of the artificial intelligent techniques which require more technically involved and mathematically complicated processes. To highlight the strengths of these approaches, the Taguchi method coupled with intelligent techniques will be employed on the fleet control of automated guided vehicles in a flexible manufacturing setting.

The ability to predict Quality of Service (QoS) of a software architecture supports a large set of decisions across multiple lifecycle phases that span from design through implementation-integration to adaptation phase. However, due to the different amount and type of information available, different prediction approaches can be introduced in each phase. A major issue in this direction is that QoS attribute cannot be analyzed separately, because they (sometime adversely) affect each other. Therefore, approaches aimed at the tradeoff analysis of different attributes have been recently introduced (e.g., reliability versus cost, security versus performance). In this chapter we focus on modeling and analysis of QoS tradeoffs of a software architecture based on optimization models. A particular emphasis will be given to two aspects of this problem: (i) the mathematical foundations of QoS tradeoffs and their dependencies on the static and dynamic aspects of a software architecture, and (ii) the automation of architectural decisions driven by optimization models for QoS tradeoffs.

Importance-performance analysis (IPA) is a popular prioritization tool used to formulate effective and efficient quality improvement strategies for products and services. Since its introduction in 1977, IPA has undergone numerous enhancements and extensions, mostly with regard to the operationalization of attribute-importance. Recently, studies have promoted neural network-based IPA approaches to determine attribute-importance more reliably compared to traditional approaches. This chapter describes the application of back-propagation neural networks (BPNN) in an extended IPA framework with the goal of discovering key areas of quality improvements. The value of the extended BPNN-based IPA is demonstrated using an empirical case example of airport service quality.