Production planning mid smoothing models have been analysed in the operational research literature for some time. Application of these models in industry is growing, mid they have proved to be economically beneficial. Still, we find there exist many questions concerning the length of time over which production should lx) planned: (1) Does an optimal planning horizon exist? (2) If it does, is it of reasonable length as to be useful to the practitioner? (3) What factors affect the length of the planning horizon?