Shurong Tong

Review on Application of Data Mining in Product Design and Manufacturing

Implementation of data mining technology in some areas such as banking, finance, marketing, insurance, health care, etc. has given very good results. However, applications of data mining in product design and manufacturing are not as broad as expected and many challenges are still ahead. This work analyzes the reasons for the limitation of data mining application in manufacturing industry and focuses on reviewing the state-of-the-art of the applications of data mining in product design and manufacturing. Some points of views are also discussed and finally this paper is concluded.

Analysis of data quality and information quality problems in digital manufacturing

This work focuses on the increasing importance of data quality in organizations, especially in digital manufacturing companies. The paper firstly reviews related works in field of data quality, including definition, dimensions, measurement and assessment, and improvement of data quality. Then, by taking the digital manufacturing as research object, the different information roles, information manufacturing processes, influential factors of information quality, and the transformation levels and paths of the data/information quality in digital manufacturing companies are analyzed. Finally an approach for the diagnosis, control and improvement of data/information quality in digital manufacturing companies, which is the basis for further works, is proposed.

Developing a systematic method for constructing the function platform of product family

A systematic method to construct the function platform for a specific product family is proposed. Based on the function basis and heuristic method, function modules of each product variants existing in the current product family can be identified. These modules as well as their instances can be categorized according to the relevant design variables. Then two distinct strategies, titled as multi-object optimal selection and similarity-based method separately, are adopted to identify function modules which can be shared among different product variants and can be sequentially used to construct the product platform. As a result, this mixed method can support platform identification for various product family comprised of different kinds of product variants.

Design process reengineering and planning based on modularization

Process modularization is one of the approaches to reengineer design process in order to improve the efficiency of product design and shorten design cycle. This paper, firstly, proposes a model of design process unit (DPU), which is the minimum design activity unit with three attributes, namely state, resource and constraint. Then, a general description of DPU is illustrated and, according to their attributes, similarity degrees between these DPUs are calculated with cosine function method. By similarity matrix and modularization idea, those DPUs with the same or similar attributes can be reengineered into one module, hence they can be executed concurrently during design process. Finally, an example is presented to validate the feasibility of the method.

A value-focused approach for establishing requirements’ specification of commercial aircraft

Although systems engineering processes and standards are widely used in aircraft development programs, traditional requirements’ engineering practice for commercial aircraft does not explicitly address value perceptions and associated information. In this paper, a value-focused approach is proposed to promote a better understanding of customer-value perceptions and their derivation among different levels for value-based requirements engineering of commercial aircraft. The approach is a four-step process starting from initial customer statements to a customer-value model and leading to a system-value model with associated component-value models. A set of theories and methods are introduced in order to resolve different aspects of the approach regarding the appropriate understanding of customer-value perceptions and the establishment of the value-based requirements’ specification. A case study is used to demonstrate the transformation of airlines’ initial expectation statements into three types of value models. There are two significant benefits of this approach: (a) perceived customer value can be explicitly modeled, simulated, and derived into different levels of the system development and (b) the value model can be subsequently utilized reactively for design evaluations and proactively for design optimization to generate creative design alternatives.

Reuse-Oriented Information Model of Product Design Process Module

To deal with the problem of product design process reuse, with the thought of modularization and based on the configuration technology, an information model of product design process module is proposed. This paper, firstly, presents the concept of design process module(DPM). Then, a three-layer structure model of DPM is introduced, including module layer, activity layer and parameter layer, further the description of activity model of DPM is put forward from the perspectives of design specifications and design resources. For the availability of hierarchy description of DPM in design process, a DPM structure tree is constructed, and the DPM module classes are described with the object-oriented methods. Finally, an implementation framework on the basis of DPM is developed to support the reconfiguration and reuse of design process.

Design Knowledge for Decision-Making Process in a DFX Product Design Approach

Product design is a knowledge-intensive process and involves large quantities of decisions. The efficiency and effectiveness of these decisions depend on the provision of many kinds of related knowledge to designers from different sources throughout the product lifecycle. First, this paper makes a literature review about the related works on product design knowledge and classifies the product design knowledge into three categories, namely product, process and product support knowledge. Secondly, the knowledge needs of product design decisions are analyzed. The decisions made during the design process have a critical impact both on the design solution obtained but also on the design process itself. These decisions are of many different types and made by different kinds of engineers and managers. This work illustrates decision-making in product design and some related works are reviewed. The decision-making during the product design process is analyzed and classified into two main categories, namely organizational and technical decisions, which are illustrated in a three dimensional figure. Finally, some conclusions and future works are discussed.

Application of Data Mining in Manufacturing Quality Data

Knowledge on product quality is one of the most important knowledge sources throughout the product lifecycle for the efficiency and effectiveness of product design decisions. To provide quality related knowledge, this paper proposed one data mining based knowledge discovery approach. This approach can extract knowledge on product quality from large volume of quality related manufacturing data. The effectiveness of this approach is illustrated and validated by an example adapted from literature. Finally, some conclusions and future works are discussed.

Manufacturing Quality Information Classification based on Group Technology and Quality BOM

Finding information from manufacturing quality information to support product design decisions is necessary for improving design quality. Based on the review of group technology and quality bill of material, the classification and coding of manufacturing quality information is proposed in which combines the manufacturing quality information families with part families. Then the application of classified and coded manufacturing quality information is illustrated for further use to support design decisions

Design process gene in product development

Process of product design bears analogies with the process of biological evolution. Based on this viewpoint, this paper introduces the notion of biologic heritability and mutation into product design process management. The paper presents the concept of design process gene and also discloses that the structure of design process gene is made up of an ordered sequence of pairs of design process bases. The composition of four design process bases and six pairs of design process bases are analyzed in detail.