A New Method of Measuring Safety Performance Will Soon Affect the Whole Industry

Abstract

Proposal Safety performance is an increasingly important factor in deciding which bidders get contracts. Companies also make safety performance an important element in selecting employees for promotion, or for termination. The stakes are high, and it is critical that safety performance measurements accurately reflect probability of future accidents. Conventional measurements are proving to be unreliable indicators of the future and can cause contracts to go to bidders with the highest, not lowest, probability of accidents. They punish supervisors and companies for things they cannot control. Additionally, they can unfairly derail the careers of capable employees at all levels. This new system rewards supervisors based on corrective actions entirely within their control. It receives strong approval from all management levels, and is proving much more effective in preventing accidents. Now, a new metric is available that eliminates these problems. It is based on a simple, objective, highly reliable system of evaluating corrective actions. This new system generates a single number directly indexed to future safety performance. The system eliminates problems with conventional frequency and severity ratios, and it is proving to be a powerful management tool in many ways. It is currently being tested and implemented by large and small companies throughout all of the industry. Introduction Conventional safety metrics are variations of frequency and severity calculations that summarize the consequences of accidents rather than summarizing the accidents themselves. These summaries are widely assumed to forecast the future - an invalid assumption for many reasons. As a result, people or entire companies may be punished or rewarded for things they cannot control. For example, assume two crews have identical accidents, but the first crew suffers no injuries and the other crew experiences a fatal injury. Current metrics effectively punish the second crew much more than the first, although injury severity was purely a matter of chance, and was outside the supervisors' control. This method is routinely applied in awarding contracts worth billions of dollars annually. This method of reward and punishment produces a powerful incentive for deception, ranging from creative accounting to falsifying accident statistics. Currently, a total recordable injury rate (TRIR) is popular as a measure of injury frequency. Future injury frequencies are estimated by assuming they will be about the same as last year, adjusted by extrapolating any apparent up or down trends seen over the past few years. The new method of measuring safety performance - the Problem Solving Index (vPSI) - provides a simple quantitative evaluation of the corrective actions on accident reports. By quantifying the merits of corrective actions, the vPSI system has major advantages over conventional measurements. It rewards problem solving and tangible actions. It also requires a change in corporate culture and processes. If a company uses this measurement system to modify their corporate culture, management at all levels of the organization will be greatly improved and dramatically improved HSE performance will be one of the many benefits. Current Measurement Systems Safety performance is becoming more important to many large companies as a measure of a supplier's or subcontractor's qualifications. This increase in importance indicates an effort to select suppliers with the lowest probability of having future accidents. All measurements of safety performance are valuable only to the extent they forecast future probability of accidents. Conventional measurements are based on the frequency and severity of injury and are highly unreliable for this forecasting the future. Some of the currect measurements include:OSHA recordablesLost timeDAFWC (Days Away From Work Case) frequencyTRIR (Total Recordable Injury Rate)