Bidding dispersion : An investigation into a method of measuring the accuracy of building cost estimates

The thesis ultimately considers the hypothesis that a relationship exists between the dispersion of bids for building contracts and the accuracy of the cost estimate.The need to measure accuracy of the cost estimate is discussed and various methods available are reviewed.The underlying causes of estimating variability are examined in conceptual terms and a new model of the relationship between the cost predictions and the actual costs of several competitors is proposed.The relative contribution of estimating and mark-up variability to total bid variability is discussed in respect of previous research. Analysis of previously published data is presented in support of the hypothesis that the cost estimate component is the primary cause of the dispersion of the bids. The significance of this hypothesis is then considered in relation to the measurement of estimating accuracy.An exploratory study is described in which complete sets of bids for 269 United Kingdom building contracts are analysed. Various summary statistics computed from the data are presented and illustrated for several variables thought to influence bid dispersion. Variables appearing to have a systematic relationship with the dispersion statistics are isolated for further examination.Finally, the conclusions obtained from this analysis of bid variability are compared with the model proposed, a priori, for estimating variability in an assessment of the validity of the main hypothesis.