Estimating construction contingency: Accommodating the potential for cost overruns in road construction projects

The determination of a project's cost contingency is a pervasive problem as the amount that is incorporated into an estimate is invariably insufficient to accommodate a project's actual cost overrun. The cost overruns experienced in 49 road construction projects procured using traditional lump contracts are analyzed. The theoretical probability distributions are fitted to cost overrun data derived from the sampled projects whose contract values ranged from AU $0.5 and AU $97 million. Goodness of fit tests are used in conjunction with probability-probability (P-P) plots to compare the sample distribution from the known theoretical distribution. A log logistic probability density function (PDF) was found to best describe the behavior of cost overruns. The cumulative distribution function is then used to determine a realistic probability of a cost overrun being experienced from the point at which a contract is awarded. This is required so that public sector clients can determine an appropriate contingency at contract award, which takes into account the nuances that contribute to increasing project costs. The research findings and methodology adopted to determine the best fit probability distribution provides a more robust approach for governing a construction contingency for road construction projects so that they can be procured successfully.