Strategies for construction hazard recognition

157
INTRODUCTION
The dynamic, rapidly changing and sometimes unpredictable nature of construction projects continue to be responsible for unacceptable injury rates (Findley et al. 2004; Ho et al. 2000;
Mitropoulos and Namboodiri 2011). Despite recent pursuits to attain zero incident projects in the United States, the construction sector in 2012 accounted for more fatal injuries than any other industry. In fact, according to the Bureau of Labor Statistics (2013), injury rates for construction increased by 5% in 2012 compared to 2011. Past research has recognized that such high injury rates are prevalent because workers often fail to predict, identify and respond to hazards in their work environment. For example, Carter and Smith (2006) examined method statements from actual projects in the UK and found that 10 to 33.5% of hazards were not adequately identified or accessed. In another recent study, Bahn (2012) concluded that new workers were sometimes unable to recognize up to 57% of hazards in work-representative settings. Similarly our field studies revealed that workers were unable to identify and communicate more than 40% of hazards in their work environment. This is unfortunate because proper hazard recognition is fundamental to the success of any safety management initiative (Mitropoulos et al. 2005). Therefore, the findings suggest that construction workers are susceptible to being injured as a result of exposure to unrecognized hazards and unanticipated risks (Carter and Smith 2006). Having identified this critical issue, we began our study with a thorough examination of current hazard recognition methods that are generally used in construction settings. Our examination of current hazard recognition methods revealed that the methods could be classified as either being predictive or reactive. Predictive hazard recognition methods, like Job Safety Analysis (JSA), rely on the predictive ability of workers to visualize work tasks as they will be performed and their associated hazards (Rozenfeld et al. 2010). Despite its benefits, such

158 methods have several limitations, including (1) hazards imposed by other crews and adjacent activities are ignored (Rozenfeld et al. 2010), (2) the assumption that workers can often predict precisely the sequence of activities even before work is initiated is often unrealistic for dynamic construction projects (Borys 2012), (3) past research has shown that workers are poor at predicting hazards but predictive methods assume hazard recognition competency of workers Fleming 2009). On the other hand, reactive methods for hazard recognition such as lessons learned and checklists rely on past injury records to make future improvements (Behm and
Schneller 2012; Fleming 2009; Zou and Zhang 2009). Similar to predictive hazard recognition methods, reactive hazard recognition methods have serious limitations. First, the past injury databases often are not complete because of unrecorded near-misses, unreported injuries, and underreporting (Gyi et al. 1999). Second, injury records only reflect a small subset of hazardous work scenarios that resulted in injuries, and information from past incidents is not generalizable across diverse, dynamic and different settings (Rozenfeld et al., 2010). Third, these methods require that an enormous amount of information is effectively transferred through instructional methods (Fleming 2009). Having identified such serious limitations with current hazard recognition methods, the research focused on identifying, developing and novel methods to improve the proportion of hazards identified or communicated.

Download 2.75 Mb.

Share with your friends: