Agent based discrete event simulation models for mono/multi Criteria assessment of maritime ports sustainability

Abstract

Most of the world economy is based on maritime freight transport and plays a key role in international trade, being a determining factor in the economic growth of countries. The steady and continuous increase in the demand for maritime freight transport in recent years (even in times of pandemic) brought consequences such as delays in operations due to congestion inside the terminal, congestion in the connection with the land transport network, unsuitable storage of containers, or due to human factors since these terminals operate 24/7 and it is necessary to manage work shifts to avoid unfortunate events. On the other hand, the functional efficiency of a container terminal is no longer the primary goal of a terminal, since different sustainability goals are becoming of great interest and concern port operators, port authorities, decision makers, politicians, citizens. The considerable increase in container volumes has increased the concerns on: (i) the global environmental impacts from port emissions. (ii) the environmental impacts on those urban areas which host several maritime ports because port operations can lead to environmental impacts on air, water, and land. (iii)the social concerns for the health and safety of ports workers, due to accidents still depends on a wide range of human errors as be psychological fatigue despite the automation level reached nowadays, (iv) the impacts of in/out traffic flows on the city congestion, (v) the impacts on the liveability of the areas surrounding a port. In this context, it is easily understandable why wide attention has been given by researchers to container terminals efficiency, and why terminal efficiency cannot be solely interpreted in terms of logistic efficiency, coherently with the United Nations in 2015 defined the Sustainable Development Goals (SDGs). The first part of this thesis work is focused on the modelling and simulation of a container terminal through the combination of the two main known simulation techniques, on the one hand the discrete event simulation that allows solving large problems through discretisation and the agent-based modelling that allows to incorporate different behavioural logics, which allows to simulate complex and

different technological contexts. In this way, a hybrid simulation is obtained, in which both modelling approaches usually guarantee an efficient solution in time, adaptable solutions to changes in the problems and computational stability. [...] [edited by Author]