Valuing flexible operating strategies in nickel production under uncertainty

Abstract

With increasing global demand for nickel (which is a key component of stainless steel) the focus of mineral industry is currently on the abundant low-grade nickel laterite reserves. The extraction of nickel from the low-grade laterites is a technically difficult and expensive process and, as a result, the profitability of nickel production projects is highly affected by uncertainty over future market conditions. The project value can be increased by utilizing flexible operating strategies in response to changing future market conditions and Real Options analysis provides a suitable tool for optimizing flexible operating strategies over a long planning horizon in the face of uncertainty. This paper presents the first study on the valuation of flexible operating strategies in a realistic nickel laterite production system under uncertainty of nickel price and exchange rate. In this paper, the production of ore from the three hypothetical nickel laterite mines being fed to a central processing facility is studied. The common features of nickel laterite production, such as a two-fraction (limonite and saprolite) structure of the laterite ore body, layering of each fraction, with different ore grades (concentration of nickel) in each layer, and a simultaneous mining of several ore bodies, are incorporated into the model. It is common in the minerals industry that the ore from each mine is blended to provide as constant a feed rate and grade as possible to the processing facility. However, such a constant feed strategy may not generate the best financial return. In this paper, we investigate whether higher returns can be achieved by adopting a flexible strategy of switching, at prescribed intervals of time, between different feed rates of ore from the three mines that have different quality of nickel laterite. Such flexible strategy allows the operator to change the production rate of nickel in response to changing projected market conditions. In this paper, we use an approximate stochastic dynamic programming framework in the form of the Least Squares Monte Carlo (LSM) method, which we extend to multiple switching options problem that incorporates complex features of nickel laterite production. In addition, an approach that combines a genetic algorithm (GA) with the Monte Carlo simulations is developed for preliminary assessment of options and for estimating the upper bounds on the strategy values. We compare the value (in terms of the expected discounted cash flow) of the optimal profit-maximising switching strategy for 10 year planning horizon with the NPV value of a constant feed strategy, commonly used in the mining industry. Numerical results show that the flexibility to selectively blend the ore from each mine in response to projected market conditions considerably increases the expected cash flow and the probability of larger profits, while decreasing the probability of smaller profits.