ABSTRACT
The cumulative maintenance cost can be many times the acquisition cost in case of asset intensive industries—mining, petroleum, utilities, etc. In most public services (water, natural gas, electricity, etc.) the stock of assets belongs to Regulatory Agencies, but not to operators. Companies can make use of them in order to fulfill contractual clauses of reliability, availability, cost, safety, etc. in the delivery of services to the population. Thus, most managers must face the problem of constrained budget for operating cost (OPEX) and at the same time requirement of cost and reliability indexes. In this case, the simplest managerial option is the replacement of current assets for new ones. In this paper we propose a model for asset replacement considering cost minimization and control over the risk of failure.
The model for asset replacement depends on economic and risk considerations. The economic model depends on variables such as (a) maintenance cost on yearly basis over time, (b) acquisition cost, (c) opportunity cost of capital, (d) loss in market value of the asset over time and (e) depreciation benefit. From these variables, the focus is on the estimation of time to replace asset such as annual cost is minimum— this is also called economic life. But, if managers search for a policy based on cost minimization, risk of failure can be high. Then, the risk estimation model allows the estimation of asset’s probability of failure at the time of replacement. The economic life depends on the maintenance cost on yearly basis, that is, the higher the maintenance cost the shorter is economic life and vice-versa. But, lower maintenance cost also implies in higher risk of failure. Then, for those assets with high risk of failure their economic life is expected to be short. These relationships have been explored in this paper using Monte Carlo simulation of the asset based on reliability and cost modeling.
This model has been applied to a utility company in the City of São Paulo and results indicate a waste of money. There are cases where asset are still in operation for 20, 30 years, but should be replaced in intervals of around 12 years. In other, replacement was carried out too early for risk control reasons, but that has happened at a very high cost and without the expected increase in reliability. This model has been implemented in the natural gas distribution system and managers have much more information for planning future values of indicators such reliability, return over asset, return over equity, etc.
