China 
Africa 
Explore chapters and articles related to this topic
Decision Making Under Uncertainty
Published in Charles Yoe, Principles of Risk Analysis, 2019
Imagine that STC had the ability to consult a seer who could eliminate the uncertainty about which state of nature would be realized. What would that perfect information be worth to STC, that is, what is the expected value of perfect information? Frequently, additional information is available that can improve the probability estimates for the states of nature (or the size of the consequences if they are uncertain). The expected value of perfect information (EVPI) is the increase in the expected outcome, in this case, profit, that would result if one knew with certainty which state of nature would occur. The EVPI provides an upper bound on the expected value of any additional sample or survey information undertaken to reduce the uncertainty about which state of nature is more likely.
Decision making under uncertainty
Published in Charles Yoe, Primer on Risk Analysis, 2019
Imagine that STC had the ability to consult a seer who could eliminate the uncertainty about which state of nature would be realized. What would that perfect information be worth to STC, that is, what is the expected value of perfect information? Frequently, additional information is available that can improve the probability estimates for the states of nature (or the size of the consequences if they are uncertain). The expected value of perfect information (EVPI) is the increase in the expected outcome, in this case, profit, that would result if one knew with certainty which state of nature would occur. The EVPI provides an upper bound on the expected value of any additional sample or survey information undertaken to reduce the uncertainty about which state of nature is more likely.
Analysis of the pavement deterioration uncertainty scenarios on pavement maintenance and rehabilitation planning optimization
Published in Structure and Infrastructure Engineering, 2023
Amirhossein Fani, Amir Golroo, Mohammadsadegh Fahmani, Hamed Naseri, Fereidoon Moghadas Nejad
Wait-and-See model assumes that the decision maker is able to wait till the uncertainty is over, before calculating the optimal solutions. Therefore, this approximation is based on the perfect information along the horizon planning. It could be computed by solving the optimization problem for each scenario, one by one and taking the mean value of all the deterministic solutions. Then, the problem can be defined as follows: in which is the optimal solution of the model corresponding to the sth scenario. Expected Value of Perfect Information (EVPI) which estimates the maximum amount a decision maker would be ready to pay in return for complete information in future or explains the loss of profit due to the presence of uncertainty. A relatively large EVPI means that incomplete information about the future may prove costly and it is critical to consider uncertainty.
A progressive hedging approach for large-scale pavement maintenance scheduling under uncertainty
Published in International Journal of Pavement Engineering, 2022
Amirhossein Fani, Hamed Naseri, Amir Golroo, S. Ali Mirhassani, Amir H. Gandomi
Moreover, the Expected value of perfect information (EVPI) is another essential indicator assessing the benefits of uncertainty consideration. The EVPI computes how much it is logical to pay in order to achieve perfect information about the future. EVPI is calculated based on and ‘wait and see (WS)’. The decision-makers have to wait before making a decision so as to obtain all information based on WS. That is to say, the latter is assessed by detecting the optimal solution after observing each possible realisation of the deterioration rate and budget scenarios with certainty and calculating the average value of all the deterministic solutions. The WS solution’s expected value is compared with the stochastic model optimal objective value. The EVPI is calculated based on Equation (18). EVPI estimates the value of knowing future with certainty and calculates the maximum amount a planner would be ready to pay to get complete information about the future in advance. If EVPI is an enormous value, the value of the information about the future is high and the impacts of uncertain parameters on the solution are considerable (Chen and Fan 2015). The EVPI equals 15.2 = 126.4–111.2. The EVPI value is roughly 12% of that shows the considerable cost of ignoring the uncertainty.
Decomposition methods for solving Markov decision processes with multiple models of the parameters
Published in IISE Transactions, 2021
Lauren N. Steimle, Vinayak S. Ahluwalia, Charmee Kamdar, Brian T. Denton
The view of the WVP as a two-stage stochastic program lends itself to measures of the impact of uncertainty in the MDP parameters. The VSS is the value added by solving the WVP given in (4) rather than solving a simpler version of the MMDP, called the mean value problem (MVP), wherein the DM solves a single MDP with parameters that are obtained by taking the weighted average of the parameters from each model of the MMDP. Another measure of uncertainty is the Expected Value of Perfect Information (EVPI), which is the expected amount that the DM would pay to know with which model of the MMDP they are interacting. These metrics allow for a better understanding of the impact of uncertainty on the performance of the DM’s decisions and how valuable it would be to obtain better information about the MDP parameters.