Tarun Bhaskar

A heuristic method for RCPSP with fuzzy activity times

In this paper, we propose a heuristic method for resource constrained project scheduling problem with fuzzy activity times. This method is based on priority rule for parallel schedule generation scheme. Calculation of critical path in this case requires comparison of fuzzy numbers. Distance based ranking of fuzzy number is used for finding the critical path length and concept of shifting criticality is proposed for some of the special cases. We also propose a measure for finding the non-integer power of a fuzzy number. We discuss some properties of the proposed method. We use an example to illustrate the method.

A fuzzy mathematical programming approach for cross-sell optimization in retail banking

We consider the problem of selecting the optimal list of customers to target for a cross-sell campaign in a retail bank. Target selection involves taking estimates of several parameters (response propensity, expected volume, expected profit from a customer, etc) and deciding on the list of customers to whom the offer should be sent such that a certain set of business objectives are met/optimized. We discuss some of the issues related to the target selection process, namely those of unreliable estimates and computational complexity of the problem. We propose a fuzzy mathematical programming technique to address these issues. The imprecise parameters and constraints are represented as triangular fuzzy numbers, while the problem of computational complexity is addressed through a group-level formulation. We use an example of a real-life cross-sell problem for a bank to demonstrate the method. We also provide some sensitivity analyses on critical resources.

A Hybrid Model for Network Security Systems: Integrating Intrusion Detection System with Survivability

Computer networks are now necessities of modern organisations and network security has become a major concern for them. In this paper we have proposed a holistic approach to network security with a hybrid model that includes an Intrusion Detection System (IDS) to detect network attacks and a survivability model to assess the impacts of undetected attacks. A neural network-based IDS has been proposed, where the learning mechanism for the neural network is evolved using genetic algorithm. Then the case where an attack evades the IDS and takes the system into a compromised state is discussed. We propose a stochastic model which enables us to do a cost/benefit analysis for systems security. This integrated approach allows systems managers to make more informed decisions regarding both intrusion detection and system protection.

Marketing Optimization in Retail Banking

In this paper, we address the problem of making optimal product offers to customers of a retail bank by using techniques including Markov chains, genetic algorithms, mathematical programming, and design of experiments. Our challenges were large problem size, uncertainty about estimates of customer responses to product offers, and practical issues in training and implementation. The solution had an estimated financial impact of around

An Optimization Model for Personalized Promotions in Multiplexes

20 million; it also provided other intangible benefits, including structured decision making, the capability of performing what-if analysis, and portability to other markets and portfolios.

A parallel genetic algorithm for propensity modeling in consumer finance

This article proposes a structure for personalized promotions in multiplexes using RFM (Recency, Frequency, and Monetary Value) Analysis, predictive modeling, and optimization at different stages. The stiff competition in the multiplex industry in India makes it essential for the players to have a good loyalty program. One effective way for increasing loyalty would be to introduce personalized promotions for the customers. This would make the promotion more targeted towards the needs of the customers and would keep them involved. One major challenge in implementing any personalized promotion is to have a good and efficient structure for the same. The proposed structure is capable of incorporating business constraints and providing useful business insights to help the multiplex have an effective loyalty program.

A metric for customer lifetime value of credit card customers

We consider the problem of propensity modeling in consumer finance. These modeling problems are characterized by the two aspects: the model needs to optimize a business objective which may be nonstandard, and the rate of occurence of the event to be modeled may be very low. Traditional methods such as logistic regression are ill-equipped to deal with nonstandard objectives and low event rates. Methods which deal with the low event rate problem by learning on biased samples face the problem of overlearning. We propose a parallel genetic algorithm method that addresses these challenges. Each parallel process evolves propensity models based on a different biased sample, while a mechanism for validation and cross-pollination between the islands helps address the overlearning issue. We demonstrate the utility of the method on a real-life dataset.