How can this employee realistically predict accidents next year?

• A. Use a single-variable forecast
• B. Use a moving average
• C. Use a multiple regression equation
• D. Use expert judgment

Answer: (C)

The idea is to use a model that connects several factors to the number of accidents, so the forecast reflects how different conditions together push risk up or down. A multiple regression approach does exactly this: it estimates how each predictor—such as workforce size, hours worked, safety training hours, prior accident rate, season, and the presence of safety programs—affects the expected number of accidents, while holding other factors constant. With historical data, you compute coefficients that quantify the impact of each factor, then plug in predicted values for next year to generate a numeric forecast. This lets you see not only the overall expected accidents but also how changes in staffing, training, or safety interventions would change that forecast. It also supports scenario planning, like evaluating how a new safety program or different workload levels would alter risk.

Other methods don’t capture this multi-factor influence as effectively. A single-variable forecast relies on one predictor and misses other important drivers. A moving average uses past accident counts without accounting for changes in conditions or interventions. Relying on expert judgment introduces subjectivity and lacks the data-driven precision needed for planning.