Artificial Intelligence for Logistics
Driven by a number of research projects worked on over the past years, my past and current work has many “data-driven” components. There are many examples of research output I worked on, which fit the dimensions descriptive analytics, predictive analytics, and prescriptive analytics. Data-driven logistics research is becoming more and more important following ever growing rich datasets. I will build more extensively upon this evolution.
In a recent OECD report, the Data-Driven Innovation (DDI) cycle is discussed in details. DDI is described as a sequence of phases from datafication to data analytics and decision making.
Datafication and data collection involves data generation through the digitization of content, and monitoring of activities. The result of datafication and data collection, can be seen as “Big Data”. This last needs to be exploited through advanced data analytics, through visualization, analysis, etc. Over time, a knowledge base evolves, i.e. the state of the learning system (e.g. machine learning). The value of big data and the knowledge base is explicitly exploited through decision making (actions). Decisions taken in turn lead to more or different data and are the start of a new cycle.
Within this described framework, my future research focus will be more and more on the interface between knowledge base and decision making, or on the prescriptive analytics dimension. Of course, still with a focus on urban logistics and multimodal networks. The understanding, application and optimization of this connection between data/information and decision making is an extremely exciting research area, especially applied to a number of domains in which I have good expertise, i.e. urban logistics and multimodal networks.
OECD (2015), Data-Driven Innovation: Big Data for Growth and Well-Being, OECD Publishing, Paris. http://dx.doi.org/10.1787/9789264229358-en