A group decision making method to manage internal and external experts with an application to anti-lung cancer drug selection

Abstract

With the changes of lifestyle and environment of people, the incidence rate of lung cancer has increased year by year, and lung cancer has become one of the most malignant tumors that threaten the health of people. Within this context, choosing appropriate anti-lung cancer drugs is of great significance for the treatment of lung cancer patients. To improve the accuracy of anti-lung cancer drug selection, it is necessary to invite many experts to participate in the evaluation process, and such a selection process can be regarded as a large-scale group decision-making problem. In existing group decision-making models, there are two hypotheses: one assumed that all experts are independent, while the other assumed that experts have certain relationships. However, in practical decision-making problems involving both internal and external experts, it is common that only some experts have mutual relationships. To address this issue, this paper proposes a large-scale group decision-making model considering the trust relationship between a set of experts. We divide experts into internal experts and external experts. The internal experts are supposed to be not independent of each other due to trust relationships, and we analyze the relationships between internal experts through the DEMATEL method. The external experts are supposed to be independent of each other. Considering the non-cooperative behaviors of experts, we provide a confidence-based adaptive consensus reaching mechanism for internal experts and a delegation-based adaptive consensus reaching mechanism for external experts. The two expert panels reach consensus through their separate consensus reaching mechanisms, and the moderator determines the optimal alternative by combining the final opinions of the two expert panels. Finally, an illustrative example about the selection of anti-non-small cell lung cancer drugs is presented to show the validity and practicality of the proposed model.