- Author
- Carter, G. | Ignall, E.
- Title
- Simulation Model of Fire Department Operations: Design and Preliminary Results.
- Coporate
- Rand Institute, New York City, NY
- Report
- R-632-NYC, December 1970, 40 p.
- Distribution
- Available from National Technical Information Service
- Keywords
- simulation | fire departments | operations research | fire fighting | response time | effectiveness
- Identifiers
- some background on New York City Fire Department operations; generating policy options; global vs. internal effectiveness criteria; edge effects; rare events-virtual measures
- Abstract
- This Memorandum describes a simulation model designed to compare different policies for locating, relocating, and dispatching fire-fighting units. Simulation experiments were made to compare proposed solutions to workload and response problems to the then current policy in New York City. The proposed policies involved creating new units, some of them to work only in peak alarm rate hours (Tactical Control Units) and reducing the nominal number of fire fighting units dispatched to street box alarms during those hours (Adaptive Response). We report in this paper a simulation experiment in which one new policy reduced both the total number of responses made by all engines and the average time from first report of the fire to the arrival on the scene of an engine and, if needed, a second and a third engine. This and ohter simulation experiments are credited with assisting in the adoption and implementation of TCU's and AR in New York City in late 1969. Several issues concerning the design and use of the simulation model are described. The first of these is the use of internal measures of performance, such as average time from first report to arrival of an engine, in place of global measures, such as loss of life. The second is the integrating of insights gained from analytical models of the various sub-problems (initial dispatch, relocation, etc.) into policies to be tested by simulation. The third is the use of "virtual" measures, a Monte Carol-like technique for efficient handling of loss of life and other important but rare events. Aspects of the operations of the Fire Department of the City of New York and their translation into SIMSCRIPT I.5 program are described. In addition, programs that provide input to that simulator and analysis of simulation output are described. Statistical analysis is made of the reliability of the differences in time required for arrival of the third engine under a proposed policy and the current one.