摘要：This article investigates personal, activity-based tour planning (TP) for tourists in public transportation networks. The objective is to propose an effective framework based on tour parameters, including starting days and times, duration, priority rates of cities, a list of activities and their duration, and a list of transportation means. The subsequent TP should consider the duration of trips to maximize the scores for visiting and acting in a given time frame. In this framework, two nested modules are of primary interest: first, in the main module, the order of the visited cities is determined using a genetic-based strategy; second, in the subordinate module, the shortest path in a multimodal network is obtained based on an adapted Dijkstra algorithm. Since TP is either personal or activity based, constraints of activity duration and periods of enduring activities are considered in the main framework. To evaluate the capabilities of the proposed framework, data are used from real transportation networks in 15 major cities in Iran. Then, several tour planners are employed in order to substantiate the performance of the proposed evolutionary framework. The resulting average relative error of the performed trials is 4.38%, and the simulation therefore vividly demonstrates that the proposed framework is suited for attaining optimum tour plans.
