Keywords: crane, construction accident, construction safety, electrocution, fuzzy logic, fuzzy set, overhead power line.

Crane accidents result in many serious and fatal injuries each year. According to the data kept by Occupational Safety and Health Administration (OSHA), crane accidents claim fifty lives in the United States of America each year. The crane coming into contact with a power line is the most common cause of fatal accidents- roughly 40% of all fatalities are attributable to electrocution [1]. Mobile cranes are involved in more than 90% of accidents that involve cranes, this is due to the fact that these types of cranes are mobile and can face a higher risk than non-stationary cranes.

Causes of electrocution as a result of crane contact with overhead power lines can be classified to three categories, triggering, enabling, and procedural causes. Enabling causes are defined as events that contribute to the deficiencies in the design, construction, or maintenance, while triggering causes are external events that could initiate failure. Procedural causes are frequently hidden events that produce both enabling and triggering events and arise from the interrelationship among various parties involved in the project.

Safe crane clear distance from an overhead power line is usually measured qualitatively and qualitatively. Quantitative measurement of the crane clear distance is related to direct measurements of the crane safe clear distance. Note that OSHA part 1926.550(a)(15)(ii) states that "For lines rated over 50 kV., minimum clearance between the lines and any part of the crane or load shall be 10 feet plus 0.4 inch for each 1 kV. over 50 kV., or twice the length of the line insulator, but never less than 10 feet" [2] In construction, specifications and standards can provide a basis for quantitative measurements of performance. In quantitative measurements, historical data of the safe crane clear distance of past empirical and analytical measurements are the basis for obtaining quantitative measurements of crane clear distance.

In practice, however, crane operators use their subjective judgment, which is imprecise yet useful, to observe the safe distance. They may think or express such a distance in terms of very safe, safe, or fairly safe, associated with very long, long, or fairly long clear distance. Such linguistic terms are not easily defined, yet essential to construction safety practice. Furthermore, when assessing the likelihood of crane contact with energized power lines, one would not necessarily use numerical probabilistic values, but rather such expressions as high, very high or low likelihood. These linguistic values can be called "fuzzy sets".

In this paper, the authors introduce a rotational fuzzy set model, whose objective is to link crane clear distance to electrocution accident probability of occurrence. Linguistic terms such as long, very long and fairly long clear distance can express the crane clear distance. Additionally, other linguistic terms such as low, fairly low and high express the likelihood of electrocution accidents as a result of direct contact of crane with overhead power lines. Approximate reasoning methods are used to relate these linguistic variables.

Suppose the following proposition is known. Given the premise, if the clear distance of a mobile crane to an overhead power line is very short (about 5 feet from the power line), the related likelihood of electrocution accident to take place is high. A question arises as to the likelihood of an electrocution accident to take place when the fact is known that the clear distance is fairly short (about 8 feet from the power line). The fuzzy rotational model developed in this paper is used to solve this problem.

In order to solve this problem, a three step process is performed. First the inverse truth functional modification (ITFM) is conducted. Then the Lukasiewicz implication rule (LIR) is applied to obtain the truth value of high likelihood of electrocution, denoted by T with a membership denoted by w. Finally, the truth function modification (TFM), that is a logic operation, which modifies the membership function of a fuzzy set with known truth values, is implemented to find the likelihood of an electrocution accident to occur.

The rotational model involves the use of ramp functions having both positive and negative characteristics. A graphical model is presented to determine the performance of a crane associated with its operation to avoid electrocution. The expected result of this study is a computer model that can be used to simulate the safety assessment of a crane operation on a construction site. The potential benefits of this tool are multifold: (1) users can use their subjective opinion to monitor crane performance, (2) they can also employ this tool as a quality control model, and (3) when using this tool, they do not need to understand the underlying concept of fuzzy logic which is often complex in nature. Hence, we expect that the study will contribute to identifying causes of electrocution accidents that take place in construction sites and avoiding the recurrence of fatalities in the future.

1

OSHA (1990). Analysis of construction fatalities-the OSHA database 1985-1989. Washington, DC: U.S. Department of Labor, Occupational Safety and Health Administration.

2

OSHA Regulations (Standards - 29 CFR): Cranes and Derricks. - 1926.550. Washington, DC: U.S. Department of Labor, Occupational Safety and Health Administration.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £135 +P&P)