Abstract: In both the public and private sectors, there is a drive to implement zero-emission solutions in traditionally polluting industries to limit their damaging environmental impact. The state of California, in particular, has implemented new legislation requiring higher percentages of freight trucks and commercial fleets to be zero- emission vehicles starting in 2024. However, the lack of medium- and heavy-duty refueling stations across California hinders the transition to zero-emission freight. Selecting the locations for such future stations must balance demand distribution, monetary costs, land resources, stakeholder feedback, and the potential for disruptions from natural disasters, cyber interference and socio-economic shocks and stressors. We solve this logistics problem near-optimally by minimizing travel time while accounting for additional variables using a generalized inexact k-medoids method. This approach is relatively fast and scales computationally such that it can be applied on a statewide scale. We propose a near-optimized network of zero-emission stations across the state of California. Our network establishes a mean travel time from a location where a driver is expending fuel to a refueling station of six minutes and minimizes diversions from existing freight routes. In this way, our facility location model can hasten the conversion from a traditional to a clean freight industry in California and can be scaled up to a national level. While facilitating the reduction of emissions from freight infrastructure, our methodology also has the flexibility to incorporate a variety of stakeholder values.