A Discussion of the Grey Model Based Methodology in Work Safety Target Forecasting and Setting

摘要

In managing occupational safety or work safety, it has been a common approach in many countries of forecasting and setting quantified and time-limited targets to reduce work-related accidents, fatalities and injuries, and this is usually included in a wide range of work safety plans and policies formulated and implemented by the government. In this paper, we present a new method for forecasting and setting safety targets by using a combination of Multivariate Grey Model and Linear Regression Model. To develop a better understanding of how to forecast the likely number of fatalities in the target year, we first perform a descriptive analysis to identify what factors within the broader economic, societal and technological backgrounds may contribute towards changes in the incidence of workplace fatalities over time, and then use seven indicators as the mark for these factors to generate a Multivariate Grey Forecast Model. Also, Accident Reduction Factors (ARFs) commonly used in road safety field are incorporated to provide a computationally simple and quick way of estimating fatality reductions, taking account the aggregated effectiveness performance of the work safety policy interventions. Finally, an application of the proposed function is made by using fatality records in several consecutive years to forecast and set safety targets for the target year. Our final objective is to explore the potential utility of the multivariate forecasting model to provide decision support for related agencies to forecast and set new safety targets and to estimate the effects of new safety measures in the target year.