Many transit agencies in North America suggest semi-flexible transit (SFT) as a viable solution to the growing demand for highly personalized and expensive paratransit services with an increasingly aging population, high operating costs associated with low demand bus transit routes, and lacking adaptability of fixed-route bus transit to serve increasingly diverse spatiotemporal travel needs. This thesis proposes an effective methodology for the optimization of SFT for operation along an under-performing low demand bus transit route in Regina, Canada. In this thesis, three research questions are addressed: (1) What levels of demand are optimal for SFT operation, given the two service delivery models, in-house transit, IHT, and contracted-out taxi, COT? (2) How to optimally design service headway (h) and slack time per trip for route-deviation (Δt) in an integrated SFT that serves both fixed-route and paratransit demand? and (3) What is the optimal vehicle size and vehicle technology for SFT operation when comparing two technologies: battery-electric vehicles (BEV) and diesel-based vehicles (ICEV)? Analytical and metaheuristic optimization techniques are employed to determine the optimal value for decision variables. The findings suggest that SFT with COT delivery model is most economical in terms of operator cost when demand is unexpectedly low, SFT with IHT delivery model is more economical when demand is low to medium, and conventional bus transit operating in-house is more cost-effective when transit demand is high. Operator cost favours solutions with low service frequency (i.e., high h), user cost favours lower ranges of h and Δt, and higher service benefit is derived from high Δt; thus, medium ranges of h and Δt appear to provide the most reasonable trade-off for service. It is also observed that for low demand (5-15 pass/hr), in terms of the total cost (i.e., operator, user, and environmental) a minivan ICEV outperforms all other scenarios as the potential savings in energy cost in favour of BEB were offset by the present high cost of installing fast chargers and when demand increases, minivan BEV outperforms. Besides contributing to the state-of-the-art research on SFT optimization, the study models are used as part of a decision support tool to establish contracting, transit network planning, vehicle technology, operation, and fare policies.