Transit networks are facing a large change due to the integration of new technology on routing and booking of rides. By reinventing paratransit and dial-a-ride services, new forms of on-demand transit services have been trialed. This, in part spurred by the proliferation of transport network companies (TNC’s) such as Uber, has led to increased demand for similar public services within cities around the world. These services give potential for the expansion of transit catchments into areas that would have received poor or no service previously, often due to lack of dense demand. Clear cost-savings have been promoted as a reason for potential service deployment as a replacement for fixed route services, with some clear, but limited, successes. The services thus far deployed have repeatedly shown to benefit local populations, especially those currently experiencing mobility poverty. As on-demand services are currently in their infancy, there is limited research currently available demonstrating a single best operational model. Current services in Australia and around the world use different booking technology, different fleets (sizes and vehicles), operate at different hours, under different pay structures, and in different environments. This variety in deployment characteristics makes it difficult to identify preferred approaches. What works well in one service may by detrimental when implemented in another. This research delivers methods of comparing these disparate services. Important factors include ease of replicability and use of publicly accessible data. Comparison techniques have to be reproducible and implementable world-wide. They also need metrics that compare performance beyond total ridership or cost, and that allow one to understand impacts for dispersed populations. The initial work in this thesis included the production of a systematic literature review on previous research examining on-demand transit services’ impacts on social equity in low population density environments. The results from this analysis found that services initially were deployed to provide access to certain disadvantaged groups throughout society. However, recent research and services expanded these disadvantaged groups to allow for increased use of the services by more of the community. Currently deployments are often open to the public with concessions provided for disadvantaged individuals. The second paper demonstrates the results of a survey of the users of the Logan DRT Trial in Queensland. The survey collected data on riders’ socio-demographic profiles, ease of access to transit, mode substitution behaviours, and access to automobility. Current user groups for this trial are similar to other on-demand services, reinforcing the benefits of these services for transport disadvantaged groups. The third paper includes an initial application of a density-based service evaluation technique. Deployed examining the natural experiment of the Logan DRT Trial in Queensland, Australia, it evaluated three different on-demand services zones with identical characteristics. Zones were ranked based on a metric which incorporated land use characteristics such as zone size and population covered as well as ridership data. This provides a simple analysis that can be reproduced across services around the world with relatively basic data. More granular analysis was also demonstrated based on zonal segmentation, demonstrating the ability to explore service performance with high levels of data for organizations with more advanced capabilities. The fourth paper uses the same metric, but applies it to a variety of service deployments with many varied characteristics. Services examined were operating in multiple Australian states, using a variety of technologies and operators, serving riders at different hours of the day, in different urban contexts, with different cost structures, and access to different destinations. By using this metric, is becomes possible to examine which services are performing well and identify attributes of success. Further argumentation is provided, demonstrating some of the benefits of further disaggregation of public data using a demonstration of the potential for statistical analyses is provided, if released data were disaggregated to match census blocks. Paper five fulfils the need of any thesis studying society published over this time period; a study incorporating the impacts of Covid-19. This paper explores the impacts on ridership of on-demand services in New South Wales throughout the initial outbreaks of Covid in the area. The study finds that urban service ridership faced larger decreases than rural services. Rural services increased in ridership since the previous year while urban services ridership fell dramatically. This contributes to the body of research on Covid-19’s impacts across different modes of transport, in different jurisdictions, during the pandemic.