Suresh K. Nair

A model for equipment replacement due to technological obsolescence

We consider the problem of deciding whether to keep a piece of equipment or to replace it with a more advanced technology. This decision must take into account both the nature of the available replacement technology and the possibility of future technological advances. Existing models are restrictive in the way they model the appearance of future technologies and the costs and revenues associated with those technologies. In an earlier paper we allowed the probability of appearance of new technologies to be non-stationary in time but required the costs and revenues of technologies to be different but constant over time. In this paper, we allow the technology forecasts and revenue functions associated with technologies to be non-stationary in time and consider salvage values for technologies. We develop a simple and efficient algorithm for finding the optimal decision using a forecast horizon approach. This approach finds the optimal decision in any period with minimal reliance on forecast data.

Inventory rationing via drop-shipping in Internet retailing: A sensitivity analysis

Abstract In this paper, we study a threshold level inventory rationing policy that is of interest to e-tailers, operating in a business to consumer (B2C) environment and selling non-perishable, made-to-stock items such as books, CDs, consumer electronics, and body and bath products. A Monte Carlo simulation model is developed to examine this policy when the demand process is stochastic, lead-time is stochastic, and the e-tailer uses ‘drop-shipping’ as an order fulfillment option. The methodology presented, which includes computer simulation and a full factorial experimental design, permits understanding of the complexity of the decision-making environment and implications of different sources of uncertainty (e.g. demand variability and lead-time variability) on a profit-maximizing threshold level of inventory, a stock level below which low margin orders are drop-shipped directly from the e-tailer’s supplier rather than fulfilled from internal stock.

Timing replacement decisions under discontinuous technological change

We consider the problem of deciding whether to keep a piece of equipment or replace it with a more advanced technology in an environment of technological change. Our model assumes that the costs associated with the presently available technology and future technologies are known, but that the appearance times of future technologies are uncertain. We develop a procedure for computing the optimal keep-or-replace decision that iteratively incorporates a technological forecast. For a certain class of situations, we show that our approach requires the minimum possible amount of forecasted data.

Infrastructure development for conversion to environmentally friendly fuel

An important concern for any nation wishing to convert to alternate, environmentally friendly energy sources is the development of appropriate fuel distribution infrastructure. We address the problem of optimally locating gas station facilities for developing nations, like India, which are in the process of converting from leaded to unleaded fuel. Importantly, a similar approach may be used in developed countries, which are in the process of converting to automobiles using hydrogen or electrical energy. An integer-programming model with the objective of balancing the perspectives of coverage and cost is presented for this facility location problem. Given the existing network of roads, the model considers the traveling population, the location of existing facilities and the cost of, either converting these facilities to carry unleaded fuel, or of installing new facilities in an attempt to minimize cost and simultaneously maximize coverage of population. We develop a heuristic solution procedure for this problem. The methodology is applied to data sets obtained from Current et al. [J.R. Current, C.S. ReVelle, J.L. Cohon, Decision Sciences 19 (1988) 490] representing the Ohio state limited access highway network, and to the Indian national highway network. Additionally, extensive simulations are carried out in order to examine where our approach compares with the maximum weighted spanning tree approach. This work extends the Maximum Covering/Shortest Path problem (MCSPP) formulated by Current et al. [J.R. Current, C.S. ReVelle, J.L. Cohon, European Journal of Operational Research 21 (1985) 189] to accommodate multiple sources and destinations. � 2002 Elsevier Science B.V. All rights reserved.

Designer-moderated product design

A well-designed product can provide a firm with a competitive advantage that will enable it to achieve a higher market share and increased profits. Conjoint analysis procedures are often used for product design. Numerous methods have been developed to take conjoint data and produce optimal or near-optimal product designs. But conjoint analysis, while significantly contributing to the product design process, still has the limitation of designing a product based only on input from the consumer. Under certain circumstances, designer preferences are an important source of expertise that needs to supplement consumer preferences in the product design problem. This research proposes a unique way of designing products using the distinct and parallel opinions from both the consumer and the designer, a concept the authors call designer-moderated product design. This paper: (1) makes a case for using designer as well as consumer preferences; (2) formulates a share-of-choices model for designer-moderated product designs; (3) develops and evaluates a heuristic to solve the problem using genetic algorithm techniques; and (4) presents a real application of the model and the heuristic.

MARKOVIAN DETERIORATION AND TECHNOLOGICAL CHANGE

Abstract We consider the classic equipment replacement problem under Markovian deterioration with the additional feature that the replacement technology is subject to change via a breakthrough. First, we examine the interaction between deterioration and technological change and derive some general properties regarding this relationship that can be used to rule out some suboptimal actions. Next, we show that although the possibility of a breakthrough does act as inducement to keep the existing technology longer in certain circumstances, this is not generally the case. Finally, we present an algorithmic forecast horizon approach to determining whether or not to keep the present piece of equipment that makes use of a finite amount of future forecasted data. We demonstrate that our approach, which is specifically designed for this problem, is both simple to implement and more “efficient” than existing general purpose approaches, in the sense that the forecast horizons generated are shorter.

A model and solution method for multi-period sales promotion design

Abstract This research addresses the optimal design of a series of promotions (which might offer free gifts, discounts, or special services) periodically mailed to potential customers. A model and methodology are presented which maximize the multiple purchases of these customers over time using opinions from both promotion designers and customers. A Genetic Algorithm-based heuristic is developed to efficiently arrive at good promotion designs, and the methodology is applied to a problem using real data.

AdPalette: an algorithm for customizing online advertisements on the fly

Abstract In this paper, we address customization and dynamic optimization of online advertisements. For online ads that attract click-throughs, we use click through rates to develop a methodology for customizing advertisements on the fly by changing content, copy, placement, animation and other attributes. We use techniques from optimization, conjoint analysis and genetic algorithms. Ads are reconstituted on the fly using graphic files for each level of each attribute, much like a painter would use a palette. We show that this approach improves response rates, reduces server storage requirements and improves ad efficiency.

Mobile Advertising in Capacitated Wireless Networks

The growing number of mobile subscribers has attracted firms to invent newer strategies to reach prospective customers in innovative but nonintrusive ways. While customer mobility creates an opportunity to reach them at desired times and locations, in practice, real-time ad deliveries are difficult due to the size of the ad-delivery decision problem. This research aims at analyzing and developing a decision policy for delivering ads on mobile devices such as cell phones. We look at the ad delivery problem from the perspective of an advertising firm, which delivers ads on behalf of its clients (merchants) to mobile customer using a wireless carriers infrastructure. We formulate the mobile ad delivery problem as a Markov decision process (MDP) model. The ad delivery policy depends on the customers desire and willingness to receive ads, their real-time locations, historical mobility patterns, available network capacity, and fee structure agreed upon for ad deliveries. We also develop a fast heuristic to solve larger size problems. We test our heuristic against an upper bound we developed and analyze performance using simulation

Managing credit lines and prices for bank one credit cards

We developed a method for managing the characteristics of a banks card holder portfolio in an optimal manner. The annual percentage rate (APR) and credit line of an account influence card use and bank profitability. Consumers find low APRs and high credit lines attractive. However, low APRs may reduce bank profitability, while indiscriminate increases in credit-lines increase the banks exposure to credit loss. We designed the PORTICO (portfolio control and optimization) system using Markov decision processes (MDP) to select price points and credit lines for each card holder that maximize net present value (NPV) for the portfolio. PORTICO uses account-level historical information on purchases, payments, profitability, and delinquency risk to determine pricing and credit-line changes. In competitive benchmark tests over more than a year, the PORTICO model outperforms the banks current method and may increase annual profits by over