Safety Wp

Submitted for inclusion in the proceedings of the 10th Annual Conference of the International Group for Lean Construction.

WORKING NEAR THE EDGE: A NEW APPROACH TO CONSTRUCTION SAFETY
Gregory A. Howell1, Glenn Ballard2, Tariq S. Abdelhamid3 and Panagiotis Mitropoulos4 ABSTRACT Construction safety has substantially improved, but has reached a plateau. Further improvement will come fromspreading Best Practice throughout the industry, or from Breakthrough that transcends Best Practice. We are working on Breakthrough and propose that what is needed is a new theory of accidents. Current Best Practice is described along with its underlying theoretical assumptions. An alternative theory is proposed, based on the work of Jens Rasmussen, a leading thinker on risk management in dynamicenvironments. A research program is proposed to test that theory and to develop a new approach to safety management. KEYWORDS accident, accident theory, decision making, hazard, risk, safety

1

Executive Director, Lean Construction Institute, Box 1003, Ketchum, ID 83340. 208/726-9989. ghowell@leanconstruction.org 2 Glenn Ballard, Research Director, Lean Construction Institute, 4536 Fieldbrook,Oakland Ca. 510/5308656. gballard@leanconstruction.org 3 Assistant Professor, 207 Farrall Hall, Construction Management Program, Michigan State University, East Lansing, MI 48824-1323. tabdelha@msu.edu 4 Ph.D., Process Improvement Coordinator, Menlo Park, CA, 94025. takism@earthlink.net

Submitted for inclusion in the proceedings of the 10th Annual Conference of the International Group for LeanConstruction.

INTRODUCTION Lean advocates minimizing waste and continuously improving. Incidents that disrupt the flow of work or lead to injury are waste, so the relationship between lean and safety is clear. Ohno’s rule that a worker stop the line rather than release a defective part downstream also resonates here. It is one thing to focus on the worker facing a hazard (a hole in the deck) andanother to require the person who completed the previous work to assure such hazards are not left open before declaring completion and allowing the next crew onto the deck. Making work flow more reliable seems an obvious way to reduce the unexpected events that lead to incidents, but so far we have only anecdotal evidence that more reliable planning does reduce incidents. We have long understood the‘soundness’ quality criterion for assignments as applied in the Last Planner System to include consideration of safety issues. A number of proposals have been put forward to make such consideration more explicit. For example, if a pre-task hazard analysis were required before an assignment was released, both inspection and root cause analysis would be improved. Inspectors who found people inhazard could take the appropriate immediate steps and then determine if the hazard had been identified in planning or not. If not, the planning should be improved and reasons for ignoring the plan could be investigated and action taken to reduce recurrence. Despite such innovations that reconceive the relationship between planning and safety, no systematic theory or practice has yet been developed,yet improving safety performance remains a high priority. In 1996, the fatality rate in the U.S. construction industry was 13.9 deaths per 100,000 workers and the injury rate was 9.7 injuries per 200,000 laborhours . By 2000, fatality rates had fallen to 12.9 and injury rates to 8.2. OSHA legislation, increased litigation, and increasing worker medical expenses and compensation insurance costs gavethe advantage to safer contractors. People at every level became less willing to accept the carnage. Construction users began to include safety performance in selection criteria and contractors began to take action. As one contractor executive put it, “No more funerals, I have attended too many. This has to stop.” With increased attention on safety, company wide safety programs became the norm....