A Discrete-Event Simulation Model for Dynamic Function Personalization in Generalized Software Packages

Generalized software packages endowed with a large spectrum of functionalities are often underutilized because users are not always aware of all the existing and newly introduced functionalities. Displaying personalized information about the package is very useful here. In an earlier work, a Markov decision model had been formulated for a demo package of ActiveX Servers and Controls to select higher utility software functions (states) for dynamically displaying them to a user. Graduate students had tested the package. The test results have helped in constructing the transition probability matrix and the holding time matrix. This paper describes a discrete-event simulation model of user behavior in such a generalized software package. The simulation results confirm the use of personalization features in generalized software packages.