The value of information (VOI) analysis has been recognized as a useful tool for measuring how much the expected monetary value can be increased as a result of an information-gathering activity. The VOI analysis is usually applied in the context of decision making under geological uncertainty, where only a relatively small number of decision alternatives are taken into account. In this paper we discuss possible applications of the VOI analysis in optimization of reservoir development under geological uncertainty. More precisely, by means of the VOI analysis, we evaluate how much an information-gathering activity provides an increase of the maximized expected monetary value, where the maximization is considered in a probabilistic sense to account for geological uncertainty. In this application we often need to deal with a quite large or even infinite number of candidate solutions within an optimization problem, which may cause trouble for an efficient implementation of the VOI analysis. After introducing a general methodology to estimate the VOI in our context, we validate our methodology through a toy problem, and moreover apply to a simple waterflooding problem. We find out that the VOI analysis can be conducted efficiently even in optimization under geological uncertainty by specializing our methodology properly.