Try Diffusion Models.

These are well-established and there are several expansions and adaptions based on the original Bass model (1969) (which itself was an adaptation of a contagion model) that can accomodate different additional information you have about competitive or industry structure, in addition to data about adoption over time.

To get you started, see: Mahajan, Vijay, Eitan Muller, and Frank M. Bass. (1990). "New Product Diffusion Models in Marketing: A Review and Directions for Research," Journal of Marketing 54.

It depends largely on how the researcher or client organization would propose to make use of the model and its outcome (you didn't say yet). The researcher also needs to operationally define the penetration event so that it can be detected and counted or measured. If penetration is, for example, the displacement of a delivery contract for a conventional vehicle by a delivery contract for an electric vehicle, then over a time interval the penetrations form a time series whose amplitude is just the count of these events per time interval. On the other hand, if hybrid vehicles are to be considered as partial penetrations, then the degree of electrical use enters into the metric. In either case, if the objective of the researcher or client organization is to predict or estimate the time to full penetration of the market, it would be straightforward to perform a time series analysis such as an autoregressive moving average. That is as much as I can offer without more information about the intent of the ultimate client and the consequent definition of the research problem.