A major challenge in credit card portfolio management is to classify and predict credit cardholders' behaviors in a reliable precision because cardholders' behaviors are rather dynamic in nature. This is crucial for creditors because it allows them to take proactive actions and minimize charge-off and bankruptcy losses. Although the methods used in the area of credit portfolio management have improved significantly, the demand for alternative and sophisticated analytical tools is still strong. The objective of this paper is to propose a multiple criteria quadratic programming (MCQP) to classify credit card accounts for business intelligence and decision making. MCQP is intended to predict credit cardholders' behaviors from a nonlinear perspective that is justifiable because both the objective functions and constraints in credit card accounts classification may be nonlinear. Using a real-life credit card dataset from a major US bank, the MCQP method is compared with popular and similar classification metho...

Classifying credit card accounts for business intelligence and decision making: a multiple-criteria quadratic programming approach

Decision support for supplier selection is a highly researched theme in procurement management literature. However applications of group decision support theories are yet to be explored extensively in this domain. This study proposes an approach for group decision support for the supplier selection problem by integrating fuzzy Analytic Hierarchy Process (AHP) for group decision making and fuzzy goal programming for discriminant analysis. In the first step, the fuzzy AHP theory with the Geometric Mean Method has been used to prioritize and aggregate the preferences of a group of decision makers. Then consensus has been developed between these aggregated priorities using the Ordinal Consensus Improvement Approach. Subsequently, the consensual priorities of this group of decision makers have been integrated with fuzzy goal programming theory for discriminant analysis to provide predictive decision support. Finally it has been shown through a case study how the integrated approach using fuzzy AHP for group decision making and fuzzy goal programming with soft constraints has been more effective as compared to an existing approach for group decision making using only AHP.

Revisiting the supplier selection problem: An integrated approach for group decision support

It is well known that data mining has been implemented by statistical regressions, induction decision tree, neural networks, rough set, fuzzy set and etc. This paper promotes a multiple criteria linear programming (MCLP) approach to data mining based on linear discriminant analysis. This paper first describes the fundamental connections between MCLP and data mining, including several general models of MCLP approaches. Given the general models, it focuses on a designing architecture of MCLP-data mining algorithms in terms of a process of real-life business intelligence. This architecture consists of finding MCLP solutions, preparing mining scores, and interpreting the knowledge patterns. Secondly, this paper elaborates the software development of the MCLP-data mining algorithms. Based on a pseudo coding, two versions of software (SAS- and Linux-platform) will be discussed. Finally, the software performance analysis over business and experimental databases is reported to show its mining and prediction power. As a part of the performance analysis, a series of data testing comparisons between the MCLP and induction decision tree approaches are demonstrated. These findings suggest that the MCLP-data mining techniques have a great potential in discovering knowledge patterns from a large-scale real-life database or data warehouse.

Multiple criteria linear programming approach to data mining: Models, algorithm designs and software development

In credit card portfolio management, predicting the cardholder's spending behavior is a key to reduce the risk of bankruptcy. Given a set of attributes for major aspects of credit cardholders and predefined classes for spending behaviors, this paper proposes a classification model by using multiple criteria linear programming to discover behavior patterns of credit cardholders. It shows a general classification model that can theoretically handle any class-size. Then, it focuses on a typical case where the cardholders' behaviors are predefined as four classes. A dataset from a major US bank is used to demonstrate the applicability of the proposed method.

/pdf/discovering-credit-cardholders-behavior-by-multiple-criteria-5eg4nkkhin.pdf

Discovering credit cardholders' behavior by multiple criteria linear programming

Credit risk evaluation using multi-criteria optimization classifier with kernel, fuzzification and penalty factors

The objective of this research is to search an alternative data mining approach that could outperform the current approaches in credit card portfolio management. A multiple criteria linear programming (MCLP) approach has been identified as the potential technology to predict the cardholders’ future behavior. The testing results on development samples showed that (i) the MCLP approach is fully controlled by the formulation; (ii) the 3000 is a stable sample size for the robustness in the separation process; and (iii) the MCLP model can be easily adapted from two-group to multi-group separation problems.

Data Mining in Credit Card Portfolio Management: A Multiple Criteria Decision Making Approach

As manufacturers continue to provide new and unique products and services for their customers, we argue that warranty menus with differentiated options are an often-overlooked unique selling point. Meanwhile, there is a huge swathe of customers who own durable products which are still in use even after the expiration of a warranty. It is in this area that we develop a research perspective which urges customers to consider warranty plans and post-warranty maintenance strategies simultaneously when making a purchasing decision. In this paper, through the lens of a consumer, we propose an integrated model of warranty and post-warranty maintenance under a ﬂexible renewable warranty. This model takes into account repair limits and multiple failure types with time-varying probabilities. As part of this, a new post-warranty maintenance strategy is designed, in which the repair time-based corrective replacement and the age-based preventive replacement are combined. The life cycle cost rate of the product is minimised by a synthetic decision-making about the consumer’s warranty purchase and post-warranty maintenance strategy. Numerical studies show that there exists a joint optimal solution, which can help the customer to develop an optimal overall warranty-maintenance strategy under the warranty menu with minimised cost rate.Highlights An integrated model of warranty and post-warranty maintenance is developed.The warranty product has minor and catastrophic failures with time-varying probabilities.A ﬂexible renewable warranty is considered for the consumer.A post-warranty maintenance strategy with the repair time limit is proposed.The optimal warranty-maintenance strategy is derived to minimise the Life cycle cost rate. 

Ming Luo

Papers

Data Mining in Credit Card Portfolio Management: A Multiple Criteria Decision Making Approach

Optimal warranty option and post-warranty maintenance strategy under a warranty menu: from a consumer perspective