Publication Type

Journal Article

Publication Date



This paper examines the capacity investment decisions of a processor that uses a commodity input to produce both a commodity output and a byproduct in the context of agricultural industries. We employ a multi-period model to study the optimal one-time processing and (output) storage capacity investment decisions---in addition to the periodic processing and inventory decisions---when both input and output spot prices as well as production yield are uncertain. We characterize the optimal decisions and perform sensitivity analysis to investigate how spot price uncertainty affects the processor's optimal capacity and profitability. Using a calibration based on the palm industry, we study (both numerically and analytically) the performance of a variety of heuristic capacity investment policies that can be used in practice. We find that if the yield uncertainty is ignored in capacity planning, then basing those plans on the average yield is preferable to basing them (as often occurs in practice) on the maximum yield. However, planning based on the average yield performs well only when the relative (processing-to-storage) capacity investment cost is high, otherwise it leads to a significant loss of profit. We also find that ignoring spot price uncertainty in capacity planning results in a relatively small profit loss. In contrast, ignoring byproduct revenue---which constitutes a small portion of total revenues---during capacity planning substantially reduces the processor's profit.


capacity management, multi-product firm, commodity risk management, spot market, agriculture, dynamic programming, processing, storage


Agribusiness | Operations and Supply Chain Management

Research Areas

Operations Management


Manufacturing and Service Operations Management

First Page


Last Page





INFORMS (Institute for Operations Research and Management Sciences)

Embargo Period


boyabatlinguyenwang_unabridged.pdf (404 kB)
Unabridged version