A Group Decision Support System (GDSS) is a decision support tool that integrates communication technologies, computer-related technologies, and Decision Support System (DSS)-related technologies, and is utilized when groups attempt to solve unstructured problems. Therefore, if a GDSS is designed with maximum consideration of the factors that influence group decision-making, it will greatly facilitate group decision-making. However, despite ongoing research on GDSS and its commercial applications, empirical research on optimal GDSS specifications for improving GDSS effectiveness remains in its infancy. Partial empirical studies have been conducted, but they have failed to yield consistent results. In this study, we used computer simulation methods to investigate which communication patterns are relatively more effective when group size and member characteristics—among the group factors identified in prior research—are varied. The findings yielded several notable conclusions. First, regardless of group member characteristics and group size, the most efficient communication pattern was the centralized Chain type. Second, when a small number of exceptionally capable individuals were present in the group, decision-making became faster. This demonstrates that when designing GDSS specifications for improving decision-making efficiency, differentiated support for leaders is necessary. Third, diversity in members' abilities had a negative effect on decision-making efficiency. A review of GDSS research to date reveals that more empirical research on optimal GDSS specification design is needed. Therefore, investigating the specific GDSS specifications suitable for the nature of the problem, member characteristics, and other group factors can serve as a starting point for resolving such issues, and in this regard, this study is considered to have its own significance in laying a modest foundation.