Simplified estimating and control procedures for the offshore oil construction industry.

Since the mid 1980's, the UK North Sea oil industry has been forced to work in a regime of lower oil prices. This, together with the increase in construction and maintenance costs of oil production installations has exerted downward pressure and profit margins. As a result operators and contractors have been motivated to search for means of reducing and effectively controlling costs wherever possible.

In this thesis a series of cost models are presented for estimating and control of offshore construction and modification projects. These models are far cheaper to use, simpler, yet just as accurate as traditional estimating methods. The models embody the principle of resource significance which causes management effort to focus on those items which influence the value and accuracy of a cost estimate.

Resource significant tasks (rst's) have been identified as those tasks in a jobcard individually having labour man-hours greater than the mean labour man-hours for all the tasks. These tasks account for 35% of the total number of tasks but represent some 80% of the total man-hours in a jobcard. Total jobcard man-hours can be predicted simply by using this smaller number of tasks. Resource significant models have been developed for Pipework, Instrumentation and Electrical work using historical data. The models are further simplified by developing a set of equations which are used to replace the lengthy traditional Norm tables. When tested on virgin data, the developed models predicted the man-hours of a single jobcard with an average accuracy of 0.30% and a standard deviation of about <IMG WIDTH=10 HEIGHT=20 ALIGN=MIDDLE SRC="/maths/pm.gif">11%. Tests on traditional estimating methods and the developed models have revealed that while the developed models have similar accuracies to the traditional method, they save some 75% of the time (and therefore cost) of estimating.