TY - JOUR
T1 - Cost-benefit analysis with switching regimes
T2 - An application of the theory of planning
AU - Sheshinski, E.
AU - Intriligator, M. D.
PY - 1989
Y1 - 1989
N2 - An extension of cost-benefit analysis, which evaluates streams of future returns and costs, whether certain or uncertain, is presented which allows for internal optimization along the path, in particular, for optimization over the choice of processes at each instant, such as in the choice of alternative technologies. Application is made to the case of switching regimes, where it is possible to switch from one process to another but at a cost, such as in the choice of oil vs nuclear technologies for electricity generation. Use is made of our prior framework for planning theory, involving choices for the horizon and period and for event as well as time planning. The optimal path involves time planning in the case of certainty and event planning in the case of uncertainty. In the case of uncertainty the policy variables of the system are dependent both on time and on the state variables characterizing the system. One interpretation of the main result is that (s, S) inventory-type planning can be applied to the choice of alternative processes in the presence of discrete switching costs. Another interpretation is that of an extension of the Pigou point to dynamical systems. Three cases are treated: that of certainty, that of uncertainty, and that of learning from experience.
AB - An extension of cost-benefit analysis, which evaluates streams of future returns and costs, whether certain or uncertain, is presented which allows for internal optimization along the path, in particular, for optimization over the choice of processes at each instant, such as in the choice of alternative technologies. Application is made to the case of switching regimes, where it is possible to switch from one process to another but at a cost, such as in the choice of oil vs nuclear technologies for electricity generation. Use is made of our prior framework for planning theory, involving choices for the horizon and period and for event as well as time planning. The optimal path involves time planning in the case of certainty and event planning in the case of uncertainty. In the case of uncertainty the policy variables of the system are dependent both on time and on the state variables characterizing the system. One interpretation of the main result is that (s, S) inventory-type planning can be applied to the choice of alternative processes in the presence of discrete switching costs. Another interpretation is that of an extension of the Pigou point to dynamical systems. Three cases are treated: that of certainty, that of uncertainty, and that of learning from experience.
UR - http://www.scopus.com/inward/record.url?scp=0024862788&partnerID=8YFLogxK
U2 - 10.1016/0898-1221(89)90098-9
DO - 10.1016/0898-1221(89)90098-9
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:0024862788
SN - 0898-1221
VL - 17
SP - 1317
EP - 1327
JO - Computers and Mathematics with Applications
JF - Computers and Mathematics with Applications
IS - 8-9
ER -