This paper reviews statistical methods for analyzing output data from computer simulations. First, it focuses on the estimation of steady-state system parameters. The estimation techniques include the replication/deletion approach, the regenerative method, the batch means method, and methods based on standardized time series. Second, it reviews recent statistical procedures to find the best system among a set of competing alternatives.

Proceedings of the 2000 winter simulation conference

Stroboscope: State and resource based simulation of construction processes.

The history of simulation programming languages is organized as a progression in periods of similar developments. The five periods, spanning 1955-1986, are labeled: The Period Search (1955-1960); The Advent (1961-1965); The Formative Period (1966-1970); The Expansional Period (1971-1978), and The Period of Consolidation and Regeneration (1979-1986). The focus is on recognizing the people and places that have made important contributions in addition to the nature of the contribution. A balance between comprehensive and in-depth treatment has been reached by providing more detailed description of those languages which have or have had major use. Over 30 languages are mentioned, and numerous variations are described in the major contributors. A concluding summary notes the concepts and techniques either originating with simulation programming languages or given significant visibility by them.

/pdf/a-history-of-discrete-event-simulation-programming-languages-4k3ujhyp67.pdf

A history of discrete event simulation programming languages

Economic analyses have become increasingly important in healthcare in general and with respect to pharmaceuticals in particular. If economic analyses are to play an important and useful role in the allocation of scarce healthcare resources, then such analyses must be performed properly and with care. This article outlines some of the basic principles of pharmacoeconomic analysis. Every analysis should have an explicitly stated perspective, which, unless otherwise justified, should be a societal perspective. Cost minimisation, cost-effectiveness, cost-utility and cost-benefit analyses are a family of techniques used in economic analyses. Cost minimisation analysis is appropriate when alternative therapies have identical outcomes, but differ in costs. Cost-effectiveness analysis is appropriate when alternative therapies differ in clinical effectiveness but can be examined from the same dimension of health outcome. Cost-utility analysis can be used when alternative therapies may be examined using multiple dimensions of health outcome, such as morbidity and mortality. Cost-benefit analysis requires the benefits of therapy to be described in monetary units and is not usually the technique of choice. The technique used in an analysis should be described and explicitly defended according to the problem being examined. For each technique, the method of determining costs is the same; direct, indirect, and intangible costs can be considered. The specific costs to be used depend on the analytical perspective; a societal perspective implies the use of both direct and indirect economic costs. A modelling framework such as a decision tree, influence diagram, Markov chain, or network simulation must be used to structure the analysis explicitly. Regardless of the choice of framework, all modelling assumptions should be described. The mechanism of data collection for model inputs must be detailed and defended. Models must undergo careful verification and validation procedures. Following baseline analysis of the model, further analyses should examine the role of uncertainty in model assumptions and data.

Principles of Pharmacoeconomic Analysis of Drug Therapy

A nonparametric technique for estimating the cumulative intensity function of a nonhomogeneous Poisson process from one or more realizations is developed. This technique does not require any arbitrary parameters from the modeler, and the estimated cumulative intensity function can be used to generate a point process for Monte Carlo simulation by inversion. Three examples are given.

/pdf/nonparametric-estimation-of-the-cumulative-intensity-3o3dk2r4cb.pdf

Nonparametric estimation of the Cumulative intensity function for a nonhomogeneous Poisson process

Frequently an arrival process is acceptably Poisson, but the ar rival rate varies with time This note uses inversion to derive a time-varying Poisson generator whose rate function is continuous an

A time-varying Poisson arrival process generator

As a preliminary step toward the goal of an intelligent automated system for simulation modeling support, we explore the feasibility of the overall concept by generating and testing a prototypical framework. A prototype knowledge-based computer system was developed to support a senior level course in industrial engineer ing so that the overall feasibility of an expert simulation support system could be studied in a controlled and observable setting. The system behavior mimics the diagnostic (intelligent) process performed by the course instructor and teaching assistants, find ing logical errors in INSIGHT simulation models and recom mending appropriate corrective measures. The system was pro grammed in a non-procedural language (PROLOG) and designed to run interactively with students working on course homework and projects. The knowledge-based structure supports intelligent behavior, providing its users with access to an evolving accumula tion of expert diagnostic knowledge. The non-procedural ap proach f...

A prototype knowledge-based simulation support system

The object-oriented view is that a system is composed of interacting objects. These objects are defined so that they hide those data and procedures which implement their behavior, however they can be manipulated by invoking publicly accessible methods defined for their class. These methods are invoked through messages which objects send to one another. New objects can be constructed from others through inheritance and subclasses. Functions and operators can be extended by overloading. Many of these features are available because of dynamic binding provided by object-oriented languages. There are a wide variety of object-oriented languages and some have facilities specifically for simulation. There is little doubt that future simulation languages will incorporate more ideas from an object-oriented perspective, especially as a means of extending the language to a wider variety of applications.

A perspective on object-oriented simulation

Simulation has been applied to several important medical deci sion problems using the SLN language. The applications to end- stage renal disease, chronic stable angina, renal artery stenosis, and treatment of hypertension and hypercholesterolemia are reviewed emphasizing the modeling considerations and results. These investigations support the utility of simulation in medical decisions and the applicability of SLN to these problems. The general applicability of simulation to medical decision problems must be evaluated in the context of available knowledge and data. The simulation of medical decisions does not produce dependence among observations so prevalent in discrete-event simulations, but to the extent that arbitrary criteria such as cost- effectiveness are estimated, statistical concerns remain. Simula tion, particularly with SLN, can play an important role in medical protocol design.

Simulation of medical decisions: applications of SLN

SLN (Simulation of Logical Networks) is a new simulation language designed specifically to provide concepts and facilities needed to model and simulate logical networks. Logical networks provide a convenient and natural means of expressing decision processes when a general-purpose discrete-event simulation facil ity is unnecessary. SLN accepts a node and branch description of a logical network and automatically handles the construction and execution of the simulation. FORTRAN can be mixed with the SLN statements with few restrictions. SLN is highly efficient, simulating only logical networks and working as a preprocessor generating FORTRAN source which can be compiled, linked, and executed. The SLN library includes facilities for generating random variates, collecting and reporting statistics, monitoring simulation execution, and tailoring the simulation output. SLN is portable and designed for use in an environment lacking a modeling tradition.

Stephen D. Roberts

Papers

A time-varying Poisson arrival process generator

A prototype knowledge-based simulation support system

A perspective on object-oriented simulation

Simulation of medical decisions: applications of SLN

The simulation of logical networks (SLN)