Abstract
The economically viable process designs should be, besides other criteria, profitable over entire process lifetime, not only at present time. An improved process design can be achieved by an appropriate trade-off between product income, operating cost and investment. The distribution between the operating cost and the investment is naturally constrained by the utility, raw material and product cost. However, all these prices are varying over lifespan of the process. The appropriate trade-off can only be established correctly if it reflects the future variability and even unpredictability of these cost. The expected impact on the trade- off would tend to compensate the cost variations, for example at higher utility cost a higher investment can be economically viable in order to decrease the operating cost. Similarly, higher raw material cost would tend to increase the efficiency of the process technology, which usually results in a higher investment. Low product price have a similar influence. However, when all the prices are simultaneously considered the tendency of each separate impact can be different. The objective of this work is to optimise process design over entire process life-time by considering some provisions for future cost fluctuations.
The majority of process synthesis models have been single-period considering only fixed current cost. However, the utility, raw material and product costs are fluctuating rather quickly and the optimal process design at certain period of time is different from the design at another period. In order to consider future costs, when synthesizing processes, a multi-period mixed-integer nonlinear model (MINLP) has been developed. Different utility, raw material and product price projections, based on the historical prices, were considered due to the uncertainty of the forecast. As an objective a maximisation of Expected Net Present Value was selected in order to account for the future utility, raw material and product prices, as well as for the time value of the money.
The influence of future utility, raw materials and product prices on the process design can be considered and analysed where simultaneous impact of all those costs on the an overall process performance is therefore explored rather than isolated impacts from individual cost. Furthermore, the distribution of expenditure for investment, raw materials and utilities between the utility system and the production/separation part of the overall process can also be investigated.
The solutions of proposed optimisation methodology seem to be more robust, compared to the conventional optimisation, since it considers future variation of cost. The results of the process synthesis described yield process designs, which have higher probability to meet each time period optimally over their full life-span. Solutions obtained are less dependent on the external price fluctuations, are economically more attractive, and yet more sustainable.
