Yiliu Tu

Product design resources optimization using a non-linear fuzzy quality function deployment model

Quality Function Deployment (QFD) is a well-known customer-oriented methodology that is widely used to assist decision-making in product design and development in various types of production including highly customized One-of-a-Kind Production (OKP), batch production as well as continuous/ mass production. Determining how and to what extent (degree) certain characteristics/technical attributes (TA) of products are to be met with a view to gaining a higher level of overall customer satisfaction is a key to successful product design and development. Most of the existing approaches and models for QFD planning seldom consider the resource constraints in product design, nor do they normally take into account the impacts of the correlation among various TA. In other words, most of the existing QFD applications assume that the resources committed fully to attaining the design target for one TA have no impacts on those for other TA. Hence, the costs/resources required are usually worked out individually by linear formulation. In practice, design resource requirements should be expressed in fuzzy terms to accommodate the imprecision and uncertainties innate in the design process, such as ill-defined or incomplete understanding of the relationship between a given set of customer requirements (CR) and TA, the complexity of interdependence among TA, etc. A non-linear fuzzy model is proposed here to offer a more practical and effective means of incorporating the resource factors in QFD planning. The impacts of the correlation among TA are also considered. In the model, the resources for achieving the design target for a certain TA are expressed in a non-linear formulation of its relationship, correlation as well as interdependence with other customer requirements or TA. The concepts of the achieved attainments and planned attainments for TA, and the corresponding primary costs, planned costs and actual costs are introduced. Solutions to the non-linear fuzzy model can be obtained using a parametric optimization method or a hybrid genetic algorithm. A case study is also given to illustrate how the proposed fuzzy model and the optimization routine can be applied to help decision-makers in a company deploy their design resources towards gaining better overall customer satisfaction.

Identification of the optimal product configuration and parameters based on individual customer requirements on performance and costs in one-of-a-kind production

One-of-a-kind production (OKP) aims at manufacturing products based on the requirements from individual customers while maintaining the high quality and efficiency of mass production. This research addresses the issues in identifying the optimal product configuration and its parameters based on individual customer requirements on performance and costs of products. In this work, variations of product configurations and parameters in an OKP product family are modeled by an AND-OR tree and parameters of the nodes in this tree. Different product configurations with different parameters are evaluated by performance and cost measures. These evaluation measures are converted into comparable customer satisfaction indices using the non-linear relations between the evaluation measures and the customer satisfaction indices. The optimal product configuration and its parameters with the maximum overall customer satisfaction index are identified by genetic programming and constrained optimization. A case study to identify the optimal configuration and its parameters of window products in an industrial company is used to demonstrate the effectiveness of the introduced approach.

Rapid identification of the optimal product configuration and its parameters based on customer-centric product modeling for one-of-a-kind production

One-of-a-kind production (OKP) aims at manufacturing products based on the individual customer requirements while maintaining the high quality and efficiency of mass production. This paper presents a customer-centric product modeling scheme to model OKP product families by considering the relations between customer needs and OKP products. In this modeling scheme, an OKP product family is modeled by an AND-OR tree. In order to investigate the relations between customer needs and OKP products, data mining techniques are employed to achieve knowledge from the historical data. First, OKP products and customer requirements are grouped into product patterns and customer patterns, respectively, using a fuzzy pattern clustering method. Then, hybrid attribute reduction is carried out based on rough set theory to remove the irrelevant attributes for each product pattern. Finally, the relationships between product patterns and customer patterns are obtained. Based on the achieved knowledge, the different patterns of OKP products are modeled by different sub-AND-OR trees trimmed from the original AND-OR tree. Since only partial product descriptions in a product family are used to identify the optimal custom product based on customer requirements, the efficiency of custom product identification process can be improved considerably.

Rapid one-of-a-kind product development via the Internet: a literature review of the state-of-the-art and a proposed platform

The historical background of Internet-based product design and manufacturing systems for rapid development of one-of-a-kind (OKP) products is systematically reviewed. By reviewing the existing OKP systems and recent approaches of Internet-based design and manufacturing systems, the requirements for the next generation of OKP systems and the current techniques that can be used to implement Internet-based product design and manufacturing systems for the rapid production of OKP products will be discussed. The problems that emerged from recent developments are reviewed and sorted. The future trends of Internet-based collaborative design, decision support, manufacturing support, supply chain management, workflow management, Internet techniques for product design and manufacturing, product modelling, STEP-based data environment, concurrent engineering, etc., will also be discussed. The reviewed state-of-the-art approaches are used directly or indirectly as references for the development of a new generation OKP systems. A reference system structure for building an Internet-based integrated product development system is then proposed to facilitate rapid development of OKP products.

Economic and economic-statistical designs of an X¯ control chart for two-unit series systems with condition-based maintenance

The economic and economic-statistical designs of an X¯ control chart for two-identical unit series systems with condition-based maintenance is studied in this paper. This system is described using a five-state continuous time Markov chain. We assume that the system is monitored by an X¯ control chart to avoid costly failures. When the control chart gives an out-of-control signal, a full inspection will be conducted to confirm the actual system condition. Moreover, we assume that the system unit can be preventively replaced at a sampling epoch and must be replaced upon failure; the cost of preventive replacement is less than that of failure replacement. In addition, a random shift size is considered for two types of design. Based on these assumptions and by using renewal theory, optimization models have been developed to find the optimal control chart parameters for minimizing the average maintenance costs. A numerical example is presented to compare the two control chart designs. We also consider the effect of each statistical constraint on the decision variables and on the expected cost.

An efficient heuristic for adaptive production scheduling and control in one-of-a-kind production

Even though research in flow shop production scheduling has been carried out for many decades, there is still a gap between research and application-especially in manufacturing paradigms such as one-of-a-kind production (OKP) that intensely challenges real time adaptive production scheduling and control. Indeed, many of the most popular heuristics continue to use Johnsons algorithm (1954) as their core. This paper presents a state space (SS) heuristic, integrated with a closed-loop feedback control structure, to achieve adaptive production scheduling and control in OKP. Our SS heuristic, because of its simplicity and computational efficiency, has the potential to become a core heuristic. Through a series of case studies, including an industrial implementation in OKP, our SS-based production scheduling and control system demonstrates significant potential to improve production efficiency.

A novel approach to probability distribution aggregation

Todays business world is highly competitive and unpredictable, so effective decision-making is of primary importance. However, it is difficult to make effective decisions when sufficient information is not available, and decision-making in such situations involves a high risk of error. Conventional statistics based approaches to such problems are not effective, because in such situations decision-making is usually in the hands of a small panel of experts. However, the expert opinions can be represented by probability distribution functions. Thus, such a problem reduces to the aggregation of a set of probability distribution functions to an aggregated or consensus distribution. In this paper, we propose a new approach to address this problem. The novelties of the proposed approach include: (1) the problem is formulated as an optimization problem and (2) the overlapping area between an individual experts distribution and an aggregated distribution is taken to measure the expertise level of that expert and subsequently to determine the weight of the expert. The proposed approach in this paper is illustrated by an example reported in literature handled with the Delphi method, which also shows the effectiveness of our approach.

Modeling of Non-linear Relations among Different Design and Manufacturing Evaluation Measures for Multiobjective Optimal Concurrent Design

This research introduces a new approach to model the non-linear relations among different design and manufacturing evaluation measures for multiobjective optimal concurrent design. In this approach, different design and manufacturing evaluation measures are mapped to comparable evaluation indices. The non-linear relation between an evaluation measure and its evaluation index is identified based on the least-square curve-fitting method. The weighting factors for different design and manufacturing evaluation indices, representing the importance measures of these indices in the multiobjective design optimization, are achieved using the pair-wise comparison method. An example case study of automobile caliper disc brake design considering 3 design evaluation measures and 1 manufacturing evaluation measure is given to illustrate the effectiveness of the introduced approach.

A genetic algorithm heuristic approach to general outsourcing capacitated production planning problems

This paper addresses a dynamic capacitated production planning (CPP) problem with consideration of outsourcing. Specifically, the outsourcing problem considered in this paper has the following features: (1) all demands are met by production or outsourcing without postponement or backlog, (2) production, inventory, and outsourcing levels all have a limit, and (3) the cost functions are considered arbitrarily and time-varying. These features come together, leading to a so-called general outsourcing CPP problem. In our previous work, an algorithm with pseudo-polynomial time complexity was developed, which includes a formation of a feasible solution region and then a search procedure using dynamic programming techniques. Due to the computational complexity with such an approach, only small and medium problems can be solved in a practical sense. In this paper, we present a genetic algorithm (GA) approach to the same problem. The novelty of this GA approach is that the idea of the feasible solution region is used as a heuristic to guide the searching process. We present a computational experiment to show the effectiveness of the proposed approach.

A heuristic for adaptive production scheduling and control in flow shop production

Although research in flow shop production scheduling has been carried out for many decades, there is still a gap between theoretical research and industrial applications – especially in some manufacturing paradigms in which adaptive production control is necessary, such as in one-of-a-kind production (OKP). This paper presents an average processing time and lever (APT-LVR) heuristic, integrated with a closed-loop feedback control scheme, to achieve adaptive production scheduling and control in flow shop production. Having been implemented in a local OKP company, our APT-LVR based production scheduling and control system demonstrates a significant potential to improve the companys production efficiency and competitiveness.