Resumo:
Fleet planning and management is the subject of optimization research, where vehicle fleet
models are created to assist managers in short, medium and long-term decision making. In
steel companies, industrial vehicles are used to move heavy materials. The planning and
optimization of vehicles quantity of the fleet are important to reduce costs and accomplish
production plans. Logistics analysts are responsible for these activities, using systems
to monitor industrial vehicle operations. However, the analysis and evaluation of the
fleet utilization is still challenging, mainly due to the volume and form in which data is
generated by the systems. This is a kind of problem that is generally out of the scope
of fleet planning optimization models. In databases with high volume and complexity,
visual analysis is an alternative for extracting insights and important information with the
application of information visualization and data analysis techniques. This dissertation
proposes the investigation of visual analysis as a means of assisting in the planning of
industrial vehicle fleets. The research was conducted through the investigation of a real
context of industrial vehicle planning through the lens of Model Building Visual Analytics
and Design Science Research (DSR). As a result two artifcats were developed: a fleet
measurement model was developed to address the lack of a fleet measurement standard;
and Fleet Profile, a solution to support fleet planning based on the visual description of
the fleet. Two research cycles were realized, each one producing a version of the Fleet
Profile. At the end, an evaluation of the Fleet Profile is carried out with two cases of real
fleets based on the experience of potential users with the solution. From the analysis of
the qualitative data, it was found that the solution was able to support the evaluation of
the fleet use and other activities of the analysts from the visual description of the fleet.
