Urban population growth is estimated to exceed 50% by 2050 in today's urban spaces. Therefore, the mobility patterns of people and objects become a fundamental element for planning, control, and decision-making in multimodal transportation. The use of an agnostic system that allows us to obtain the best combination of technologies and cognitive predictive inference models covering all areas of transportation (road, maritime, and air) without programming language limitations, supported by probability distribution functions on the entropic maximization theory of complex stochastic systems as the core model that could be incorporated into a machine learning logical architecture. It allows for selecting the most efficient, harmonious, and sustainable transportation trajectory. The methodology employed is exploratory-descriptive and theoretical, based on experiences implemented in other countries, and the incorporation from the coupling of Shannon theory with Gamma distribution functions in multivariate stochastic systems for the transportation sector as an innovative contribution of this work. A representative model of an intelligent agnostic logical architecture is presented, where the integration of the multivariate system is shown, nourishing the argument in the justification of the use, and could be taken as a proposal to be developed and implemented to reduce road congestion, reduce environmental pollution, and provide a sustainable alternative. The challenge is the understanding of this intelligent agnostic system by legislators in the transport area for the implementation of “IoT” devices in each transport unit and routes for connectivity to a "brain" that receives information from other areas of transport and walkers from their devices with high-speed technology in data navigation.