Intelligent Cost Modelling

First principle estimating models are based on the premise that the estimator builds an item price from allocating resources from a library and assumes quantity and production values for each resource – in so doing arrives at a cost value for each of the assigned resources as they apply to the item of work to be performed.

This process can be termed ‘bottom-up resource cost allocation’ and is used widely in the industry. The process by which resources and their utilisations are allocated depend entirely on the estimators grasp of the work to be performed, the method used in the field, and the production that can be expected from the use of the resources chosen. With experience both at estimate stage, and later verification on site, the estimators develop a keen insight and predictive ability for subsequent estimates. With feedback from completed contracts, standards are established and the process can be used in subsequent bids.

It follows, therefore, that a fair amount of intelligence is formulated over multiple occurrences of these methods and so the resource, quantities and productions achieved are stored for further assignment.

There are, in most instances, over-arching parameters, or variables, that can be defined for work items and the methods used to price these items equally well known.

We can postulate therefore that these procedures and methods are intrinsically ‘intelligent’. With the definition of the over-arching parameters defined at bid stage, the process of estimation can be converted into an intelligent ‘top-down’ price production method.

For example. The production of Auger Bored Mass Concrete Piles may be work undertaken on a regular basis by an organisation. The over-arching parameters could be:

1. Average depth of the piles
2. Average diameter of the piles
3. Number of piles to be produced.

Instead of the estimator calling for resources from the library and allocating them to the work item, an ‘intelligent’ approach would be to achieve the entire estimation process by requesting the estimator to answer the three questions above. Using these parameters, the system now interrogates the library for Auger Bored piles and :

1. Calculates the volume of earth to be excavated for the piles (Ave Did x Ave Depth x No of piles)
2. Calculates the volume of concrete required (Ave. Did x Ave Depth x No of Piles)
3. Then assigns a grouping of resources carrying the Auger Bored Mass Concrete Piles to the quantities calculated above in the form of:

• Excavate in Soft Material using Auger Bore / Production
• Mass Concrete (Cement, Sand, Stone) / Production
• Operator for Auger Bore / Production
• Concrete Mixing Gang / Production
• Concrete Placing Gang / Production

The end result is a priced item with machine allocated resources and fully resourced Material, Labour, Equipment, Sub-Contract etc utilisations. This is a very simplified example but the methodology could be developed for larger and more complex work items. Also the procedure would depend on the variability of work undertaken by the contractor and may not be applicable in a number of cases, or for that matter, industries.