The complexity of traffic control systems, as well as the growing volume of traffic, interconnected missions types and mission demands on the operators, indicates that critical attention should still be paid to the problem of operator's cognitive workload (WL). On the other hand, the development of traffic control towards on-line measurement of cooperative aspects between humans and machines, is part of the more general need to measure the human agents “situation awareness” in industrial environment. The University of Modena and Reggio Emilia was partner of the European Artemis project “Designing Dynamic Distributed Cooperative Human-Machine Systems” (D3CoS 2011–2014) [1] to define affordable methods, techniques and tools addressing the specification, development and evaluation of cooperative systems where human and machine agents are in charge of common tasks, assigned to the system as a whole. One of the basic keys to reach an optimal human-machine cooperation is the measure of the human operator workload. In order to setup a possible method for the objective evaluation of cognitive workload we had to investigate aspects of the functional status of human operators interacting with a simulator in maritime domains. We recorded objective psycho-physiological measures: eye blinks, respiration rate and amplitude, electrodermal activity, heart rate variability, and blood pressure. They were analyzed and correlated with subjective self-assessed responses from two questionnaires: NASA-TLX and Rating Scale Mental Effort (RSME), with the aim to realize a mathematical model for classifying the operators' mental workload. The purpose of this paper is to present the methods, applied on a pilot study, that we carried out to discriminate the WL intensity, based on psycho-physiological signals alone.