This document is the report of the thesis "Normative thinking on wastewater treatment plants". This thesis was born from the interest of the author in Artificial Intelligence (A.I.). Having done all the subjects related with AJ. that the Barcelona School of Informatics (FIB) offers, I asked the teachers of my favorite ones for a thesis related with the A.I. . Ulises Cortés and Juan Carlos Nieves offered me this interesting thesis based on a doctoral thesis of environmental sciences done by Montse Aulinas [23]. The proposed work implied theoretical research, a working implementation and a real life domain to work with. I accepted without any doubt. Aulinas's thesis proposed a multi-agent based system to manage the problems caused by the industrial wastewater discharges in rivers. She discussed that, by the use of intelligent agents in the managing process of wastewaters, there could be an important increase in the quality of the river water and in the efficiency from the organizational point of view. To do that she proposed a group of agents, which would take the roles of the most important entities in the process of wastewater discharges, from industries to the agencies in charge of controlling them, in order to represent all the involved parts. It is obvious that, for the agents to be able to work rationally, they need to interact with the laws they are subject too That is the main issue this thesis deals with. Based on a real world doma in, this thesis proposes a way to make those laws to be comprehensible for agents. It will discuss a methodology for analyzing, specifying, implementing and testing those laws, in a generic way that can be applied to any normative environment. The goals of this thesis are, To obtain a generic and complete specification syntax for analyzing laws and norms, prove that specification with an implementation of reallaws applied to the given doma in and To develop a prototype where the norms implementation can be tested using a possible real scenario.