Academic articles

IRS Office of Research Headquarters measures and models taxpayer burden, defined as expenditures of time and money by taxpayers to comply with the federal tax system. In this research activity, IRS created two microsimulation models using econometric techniques to enable the Service to produce annual estimates of taxpayer compliance burden for individual and small business populations. Additionally, a Discrete Event Simulation (DES) model was developed to represent taxpayer activities and IRS administration in postfiling processes.

One of the approaches to modeling hybrid systems is to assign algebraic-differential equations describing the continuous behavior to states of state machines that represent discrete logic. The resulting hybrid state machine is a powerful concept to specify complex interdependencies between discrete and continuous time behaviors. It, however, exposes the simulation engine to a number of problems, which we discuss. The hybrid state machine based approach presented in this paper is fully supported by UML-RT/Java tool TimeBroker developed at Experimental Object Technologies.

A large class of systems being developed has both continuous time and discrete time behavior. In fact, any system that interacts with physical world falls in that class. Chemical, Automotive, Military, Aerospace are areas most frequently mentioned in this respect. To model such systems successfully and to get accurate and reliable results from simulation experiments one needs an executable language naturally describing hybrid behavior, and a simulation engine capable of simulating discrete events interleaved with continuous time processes. Additional problems arise with simulating hybrid systems in a distributed environment.

The presentation shows you how to develop simulation models, choose abstraction level and methodology.