Abstract:

Outsourcing various activities is a common business model in underground mining operations. This allows the mining company to diversify its risks and transfer some of them to the contractor. Furthermore, outsourcing to a contractor that specializes in certain activities (e.g. shaft sinking, lateral development or transportation) provides a significant cost advantage to the mine operator. On the other side of the coin, the contractor assumes a great deal of risk in terms of timing and costs. Therefore, the accuracy of cost and time estimates is vital. Bidding with an overestimated budget for a project during the tender process may result in the contract not being awarded. Underestimating project costs can create low profits and even financial losses. Similarly, there may be penalty costs associated with delays if the project is not completed on time. From this perspective, cost estimation is an important issue for a contractor.

There is currently no universally accepted cost estimation technique. Each mining project is quite different due to differences in factors such as lithology, geology, geo-mechanical properties and water availability. Data from previous projects and the judgment of the cost estimator may not be sufficient. Traditional techniques (e.g. order-of-magnitude estimates, cost estimating manuals, budget estimates and final estimates) have significant limitations for cost estimating in the mining industry. This presentation aims to show how the cost estimating process can be improved from a contractor's perspective. After exploring the practices of various organizations (e.g. Association for the Advancement of Cost Engineering, International Association for Cost Estimating and Analysis), the potential use of various machine learning techniques and a case-based reasoning model is explored. We hope this research will deliver better cost estimations so that contractors offer a more accurate bid in tendering.

Menu

Eduardo Sanchez

Biography