Towards a method of measurement and cost control for civil engineering work in the petrochemical industry

This work deals with the formulation of a hypothetical measurement and cost control model for Petrochemical Civil Engineering work. The conventional model used at the collaborating organisation is the conventional general building construction measurement model, adopted without modification. Therefore there is a presumption, rather than proof, that this model is appropriate to Petrochemical Civil Engineering cost data. The lack of theoretical development in the cost modelling field give grounds to suspect that the conventional general building construction measurement and cost control model have little empirical observation to guide their formulation. Thus it will be argued that the reasons to employ any such model in preference to any other are likely to be sociological influences of peer group paradigm practice. These sociological influences have been insufficiently recognised and studied. This thesis argues for the collection of empirical evidence upon which to build a hypothetical cost model for Petrochemical Civil Engineering work. and that this would constitute a small step towards the theoretical development of cost modelling. To this end a strategy of grounded theory is proposed, devoid of any of the "preconceived truths" characteristic of middle range theory. It shall be argued that, given the lack of strength in the theoretical component of model building epistemology, the first step should be to make the model fit the data rather than to employ a prescriptive model into whose requirements the data. must be shaped (this is argued to be an undesirable characteristic of the conventional models). For the purpose of measurement functions, statistical analyses shall identify significant and insignificant cost centres in the Petroleum Civil Engineering cost data. A hypothetical Petroleum Civil Engineering model shall be formulated whose parameters more closely reflect these observations. Validation of this hypothetical model shall be commenced by means of a survey of all its potential users. The survey solicits their judgement as to whether the model is likely to achieve its objective in practice. The reluctance of some users to believe the empirical findings will be argued to be symptomatic of the sociological basis upon which alternative cost models and techniques are frequently appraised. Given the characteristic property of empirical models, that they yield general trends but not precisely identical outcomes, discussion shall be offered of whether truly "standard" models for costs of construction projects are either achievable or, indeed, desirable. The project costs being observed are the outcomes of actions and decisions in the social world, not the outcomes of phenomena of the physical world. Therefore it is necessary to ask whether disimilarities are being modelled as opposed to similarities, and whether "universal" methods of measurement and cost control could ever fit any individual project to which they might be applied. It should be noted that the aim of this work is to formulate a model for a limited set of situations, not for all construction work; though there are areas of overlap between construction work types in different sectors of the industry. A test of the hypothetical Petroleum Civil Engineering model, in the form of tender produced used tender documents prepared according to the parameters on the hypothetical Petroleum Civil Engineering model will reveal favorable performance. The ensuing criticisms, by measurer and tenderer, of the hypothetical model shall be shown to be largely of a technical nature, indicating a tendency to appraise models on the basis of their internal consistency rather than on the basis of the epistemological arguments surrounding the formulation of their internal components. For the purpose of cost analysis and planning functions, unacceptably high variety of cost behaviour shall be found using conventional Elemental cost centres. This will be argued to be inadmissible in a cost model. The need to identity more consistent and significant cost centres will be argued. A po entially useful classification of such cost centres, developed by others, shall be identified. Investigation into their applicability to Petroleum Civil Engineering modelling shall be recommended as being worthwhile. It shall be recommended as an area for future work that a single model, at an appropriate level of abstraction, be sought which is capable of simultaneously fulfilling the measurement and cost analysis and planning functions. The conventional measurement and cost planning models are merely different levels of abstraction of each other, the latter being an over simplified abstraction from the over detailed former. That this abstraction is deemed to necessary at all is argued to be demonstrative of the fact that neither of the conventional levels of abstraction are appropriate.