Yunfeng Wang

A genetic algorithm for cellular manufacturing design and layout

Abstract Cellular manufacturing (CM) is an approach that can be used to enhance both flexibility and efficiency in today’s small-to-medium lot production environment. The design of a CM system (CMS) often involves three major decisions: cell formation, group layout, and group schedule. Ideally, these decisions should be addressed simultaneously in order to obtain the best results. However, due to the complexity and NP-complete nature of each decision and the limitations of traditional approaches, most researchers have only addressed these decisions sequentially or independently. In this study, a hierarchical genetic algorithm is developed to simultaneously form manufacturing cells and determine the group layout of a CMS. The intrinsic features of our proposed algorithm include a hierarchical chromosome structure to encode two important cell design decisions, a new selection scheme to dynamically consider two correlated fitness functions, and a group mutation operator to increase the probability of mutation. From the computational analyses, these proposed structure and operators are found to be effective in improving solution quality as well as accelerating convergence.

Genetic algorithms for integrating cell formation with machine layout and scheduling

Abstract Cellular manufacturing (CM) has been recognized as an innovative practice for companies to gain efficiency as well as flexibility under today’s small-to-medium lot and customization-oriented manufacturing environment. Among the necessary decisions for a successful CM implementation, cell formation (CF), group layout (GL) and group scheduling (GS) are the three most popular ones. These decisions are interrelated and may impact each other but they are often treated separately or as a sequential decision in prior research. In this paper, we propose a new approach to concurrently make the CF, GL and GS decisions. A conceptual framework and mathematical model, which integrates these decisions, are proposed. A hierarchical genetic algorithm (HGA) is developed to solve the integrated cell design problem. Two heuristic operators are proposed to enhance its computational performance. The results from our study indicate that: (1) the concurrent approach often found better solutions than the sequential one, and (2) with the proposed heuristic operators, the HGA procedure performed better than without them.

A Review of Data Mining-Based Financial Fraud Detection Research

Nationwide, financial losses due to financial statement frauds (FSF) are mounting. The industry recognizes the problem and is just now starting to act. Although prevention is the best way to reduce frauds, fraudsters are adaptive and will usually find ways to circumvent such measures. Detecting fraud is essential once prevention mechanism has failed. Several data mining algorithms have been developed that allow one to extract relevant knowledge from a large amount of data like fraudulent financial statements to detect FSF. Detecting FSF is a new attempt; thus, several research questions have often being asked: (1) Can FSF be detected? How likely and how to do it? (2) What data features can be used to predict FSF? (3) What kinds of algorithm can be used to detect FSF? (4) How to measure the performance of the detection? And (5) How effective of these algorithms in terms of fraud detection? To help answer these questions, we conduct an extensive review on literatures. We present a generic framework to guide our analysis. Critical issues for FSF detection are identified and discussed. Finally, we share directions for future research.

Concurrent design of cellular manufacturing systems: a genetic algorithm approach

Cellular manufacturing (CM) has been recognized as an innovative practice for companies to regain competitive advantages under todays small-to-medium lot manufacturing environment. Among the necessary decisions for a successful CM implementation, cell formation (CF) and group layout (GL) are the two most popular ones. These decisions have received extensive attention but are often treated as separate or sequential processes, despite their cross impact on each other. In this paper, we propose a new approach to concurrently solve the CF and GL problems. A conceptual framework and mathematical model, which integrates these two decisions and incorporates some important factors, such as operation sequence, part demand, transfer batch, machine capacity, and layout type, are proposed. A hierarchical genetic algorithm (HGA) is developed to solve the integrated cell design problem. The results from our investigations show that: (1) the proposed HGA procedure performs better than two other heuristics, and (2) the concurrent approach often finds better solutions than the sequential one.

An improved fuzzy clustering method for cellular manufacturing

Fuzzy c-means (FCM) has been successfully adapted to solve the manufacturing cell formation problem. However, when the problem becomes larger and especially if the data is ill structured, the FCM may result in clustering errors, infeasible solutions, and uneven distribution of parts/machines. In this paper, an improved fuzzy clustering algorithm is proposed to overcome the deficiencies of FCM. We tested the effects of algorithm parameters and compared its performance with the original and two popular FCM modifications. Our study shows that the proposed approach outperformed other alternatives. Most of the solutions it obtained are close to and in some cases better than the control solutions.

Privacy Preserving Data Mining Research: Current Status and Key Issues

Recent advances in the Internet, in data mining, and in security technologies have gave rise to a new stream of research, known as privacy preserving data mining (PPDM). PPDM technologies allow us to extract relevant knowledge from a large amount of data, while hide sensitive data or information from disclosure. Several research questions have often being asked: (1) what kind of option available for privacy preserving? (2) Which method is more popular? (3) how to measure the performance of these algorithms? And (4) how effective of these algorithms in preserving privacy? To help answer these questions, we conduct an extensive review of 29 recent references from years 2000 to 2006 for analysis.

Investigating acceptance of RFID in Chinese firms: The technology-organization-environment framework

The purpose of this paper is to study the influences of technological, organizational and environmental factors on the acceptance of RFID in China based on the technology-organization-environment framework. The concept of RFID acceptance is defined as a psychological status to intend to use RFID. The data came from a questionnaire survey in China, and 144 samples were analyzed. The analysis shows that environmental context and organizational context have the important influence on the RFID acceptance in Chinese firms. The environmental context includes customer mandate, competition pressure, and RFID vendor support. The Organization context includes IS usage, professionals of IS unit, and firm size. The characteristics of FRID technology itself are not significant for RFID acceptance.

An authentication protocol for secure and efficient RFID communication

A scalable authentication protocol is proposed for security and efficient RFID communication. The performance evaluation and security analysis has proved its advantages. Compared with other protocols, it can preserve content and location privacy, resist replay attacks and Dos attacks, and has forward security and scalable.

RFID application challenges and risk analysis

RFID application becomes the focus recently. As an automatic identification technology, RFID has many advantages over traditional one like barcode. It does bring about tremendous changes, and also face lots of barriers during investment. In this paper, we integrate technology risk, IT risk and RFID risk, develop a RFID risk factor framework with analogy method, and find out the relationship between RFID uncertainty and RFID risk. Subjective Scoring Method is adopted to assess RFID risk level. Finally, RFID risk management methods for different kinds of risks are provided.

Strongest Association Rules Mining for Personalized Recommendation

Abstract The article proposed the notion of strongest association rules (SAR), developed a matrix-based algorithm for mining SAR set. As the subset of the whole association rule set, SAR set includes much less rules with the special suitable form for personalized recommendation without information loss. With the SAR set mining algorithm, the transaction database is only scanned for once, the matrix scale becomes smaller and smaller, so that the mining efficiency is improved. Experiments with three data sets show that the number of rules in SAR set in average is only 26.2 percent of the total number of whole association rules, which mitigates the explosion of association rules.