Discrete Time Series Forecast of Accident/Injury Frequency Rates in the Army Materiel Command.
Abstract
The research report attacks the problem of providing the safety engineer with a forecasting model that could be used as an effectiveness measure for newly instituted safety programs. Two mathematical forecasting techniques (conventional exponentially smoothed and adaptive exponentially smoothed forecasting) are used to obtain a predicted accident/injury frequency rate. The accuracies of these two techniques are compared and it is found that the conventionally smoothed forecasts provide the most accurate predicting model, but are not considered accurate enough to be used as a measure for program effectiveness.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1975
- Accession Number
- ADA009317
Entities
People
- Norbert Ronald Kniepp
Organizations
- United States Army Materiel Command