Citation

Love PED, Teo P, Carey B, Sing CP, Ackermann F. Safety Sci. 2015; 79: 55-62.

Copyright

(Copyright © 2015, Elsevier Publishing)

DOI

10.1016/j.ssci.2015.05.009

PMID

unavailable

Abstract

Safety and quality performance share a symbiotic relationship. Despite the extensive amount of research that has provided quantitative assessments of the financial impact of rework, there have been limited studies that have examined the relationship between safety performance (i.e. in terms of the number of incidents arising) and rework that occurs from the issue of non-conformances (NCRs). Using an exploratory case study approach, the statistical characteristics of incidents (n = 16,885) and NCRs (n = 2885) requiring rework experienced by a tier one Australian contractor over a 31 month period are analyzed and their relationship determined. A significant association between incidents and rework NCRs was revealed (p < 0.05). The skewness and kurtosis values of monthly incidents and NCRs are computed to determine if the empirical distribution of the data follows a Normal distribution. The Kolmogorov-Smirnov, Anderson-Darling and Chi-Squared non-parametric tests are used to determine the 'Goodness of Fit' of the selected probability distributions. An Inverse Gaussian probability function is found to be the best overall distribution fit for the monthly incidents and used to calculate the probability of their occurrence. A Lognormal probability function was found to be the best overall distribution fit for the NCRs and also used to calculate their likelihood of occurring. Ascertaining the best fit probability distribution from an empirical distribution can produce realistic probabilities of incidents and NCRs, which should then be incorporated into a contractor's risk management and continuous improvement strategy.