The planning of a rail transit system is a complex process involving the determination of station locations and the rail line alignments connecting the stations. There are many requirements and constraints to be considered in the planning process, with complex correlations and interactions, necessitating the application of optimization models in order to realize optimal (i.e. reliable and cost-effective) rail transit systems. Although various optimization models have been developed to address the rail transit system planning problem, they focus mainly on the planning of a single rail line and are therefore, not appropriate in the context of a multi-line rail network. In addition, these models largely neglect the complex interactions between station locations and associated rail lines by treating them in separate optimization processes. This thesis addresses these limitations in the current models by developing an optimal planning method for multiple lines, taking into account the relevant influencing factors, in a single integrated process using a geographic information system (GIS) and a genetic algorithm (GA). The new method considers local factors and the multiple planning requirements that arise from passengers, operators and the community, to simultaneously optimize the locations of stations and the associated line network linking them. The new method consists of three main levels of analysis and decision-making. Level I identifies the requirements that must be accounted for in rail transit system planning. This involves the consideration of the passenger level of service, operator productivity and potential benefits for the community. The analysis and decision making process at level II translates these requirements into effective criteria that can be used to evaluate and compare alternative solutions. Level III formulates mathematical functions for these criteria, and incorporates them into a single planning platform within the context of an integrated optimization model to achieve a rail transit system that best fits the desired requirements identified at level I. This is undertaken in two main stages. Firstly, the development of a GIS based algorithm to screen the study area for a set of feasible station locations. Secondly, the use of a heuristic optimization algorithm, based on GA to identify an optimum set of station locations from the pool of feasible stations, and, together with the GIS system, to generate the line network connecting these stations. The optimization algorithm resolves the essential trade-off between an effective rail system that provides high service quality and benefits for both the passenger and the whole community, and an economically efficient system with acceptable capital and operational costs. Imperial College London vi Chro Ahmed The proposed integrated optimization model is applied to a real world case study of the City of Leicester in the UK. The results show that it can generate optimal station locations and the related line network alignment that satisfy the various stak...