This paper attempts to evaluate the reengineering option against replacement of a block of computers. For this purpose, a block replacement model for block of items using high-order Markov chains is developed and employed (Kilari et al., 2011). To make the model more realistic, two intermediate repairable states, i.e., minor repair state and major repair state, are introduced between the working and failure states of the system. Transition probabilities for future periods are estimated by spectral decomposition in First Order Markov Chain (FOMC) and by moving weighted transition model for Second Order Markov Chain (SOMC). With these probabilities, the number of systems in each state and the corresponding maintenance cost are computed accordingly. The predicted inflation for computer and computer-based system in India and the real value of money using Fisherman’s relation are employed to study and develop the real-time mathematical model for group replacement decision making, which is useful for large IT companies and any other block of items as well. In continuation to this, an attempt is made to evaluate the option of reengineering the computer networking process to reduce the cost of hardware, maintenance and support. For this thin-client technology is used in lieu of replacement option. For this, the developed block replacement model using SOMC is extended to accommodate the reengineering cost.

The thrust on the need for effective decision making in production and service sectors paved the way for the development of several mathematical models in the face of reasonable degree of certainty. However, in reality, there are several kinds of decisions to be made by taking into account the great uncertainty about one or more future events. As a remedy, stochastic models are tailored to represent, to an extent, the complexity of the real situations and the uncertainty prevailing around these situations. It is a well-known fact that it is difficult to develop a model that represents the reality as close as possible and simple for analysis. Consequently, different models each representing one or more parameters associated with real-life situations are developed. The present study is a contribution to the development of mathematical and stochastic models that help in evaluating block replacement decisions for a block of items under the influence of macroeconomic variables and determining an age at which the replacement is economical.

The replacement decisions in MNCs in ICT area are predominantly with the computers and computer-based system. The primary decision is generally whether to replace the existing computer-based system consisting of a large number of computers or to use it for some more period of time.

Several researchers investigated the optimal age-replacement models with repairs to reduce the cost. Nuthall et al. (1983) studied the impact of inflation on replacement costs along with the impact of some other parameters, viz., financing method and increased or decreased hours of use. Chein and Chen (2007) presented an agereplacement model with minimal repair based on cumulative repair cost limit. In this, they considered the complete repair cost data in order to decide whether to repair the unit or to replace it. Bagai and Jain (1994) discussed optimal replacement time under the age-replacement policy for a system with minimal repair that involves the replacement of only a very small part of the system. Rupe and West (2000) explored the maintenance models for finite time missions by considering the net present value of costs.

There are some studies on the replacement decisions for warranted products. Zuo et al. (2000) discussed replacement policy for multi-state Markov deterioration of machines that are under warranty. Yue and Thomas (2010) extended the work of Zuo et al. (2000) by considering more general state space with time parameters at each state.

Archibald and Dekkar (1996) studied and compared the optimal age-replacement, standard block replacement and Modified Block Replacement (MBR) policies with an inference that MBR policy is appreciably better than the remaining two.

However, there is not much literature on block or group replacement model with Markov chain transition probabilities that is used in many applications.

Operations Management Journal, Relative Efficiencies of Schools, Data Envelopment Analysis, Government-Aided Schools, Linear Programming Model, Decision-Making Units, Organizational Units, Human Resources, Public Procurement Sectors, Government Schools, Education System.