Big data analytics for cold chain logistics optimization in refrigerated trucks.

Data analysis to predict the demand of goods, optimise the routes in real-time and provide visualizations and descriptions.

Odin Solutions (OdinS) is a SME founded in August 2014 and accredited as an innovative ICT company (EIBT) by MINECO and ANCES. OdinS has a strong background in the R&D fields of Internet of Things, Security and Data Analytic. The pilot contributes to the development of the supply chain 4.0, specifically the cold chain, and it is aligned with the interest of OdinS to contribute in the emergence of smart environments.


Odin managed to provide a solution that shows the following information:

  • Ranking of the products
  • Trip Duration statistics
  • Seasonality of products’ demand
  • Geographic Representation of the trips

We have indexed the data using ElasticSearch and connected our dashboard to it using the elasticsearch package. 

We provided daily and weekly predictions on the demand of each of the products for the management of the products. 

We have studied and detected anomalies in fuel consumption and in trip durations in order to do predictive maintenance and in order to optimise the trucks’ routes


We used the Deep Hybrid DataCloud for deploying a Jupyter instance in the DEEP CLOUD testbed with a GPU.

This served to ease the machine learning models training that were tested for the prediction of the demand of different products (multivariate approach) and in the detection of anomalies by means of univariate predictive algorithms.

We also received support from the Services of Scientific Data Platforms Department of the Poznan Supercomputing and Networking Center in order to fasten up the descriptive analysis that is shown in our Dashboard. Following their advice we used the elasticsearch solution, that was deployed in one of their machines, for data indexing. We were able to connect our dashboard to their machines by means of ssh tunneling and the computing speed of the results was greatly improved


Nowadays, many businesses are concerned about collecting data and make great efforts by deploying sensors. However, they lack solutions that extract information from such data. The transportation sector is not an exception.

Our solution provides meaningful knowledge for the logistics business, by giving a better understanding of the trips : product’s groupings, duration and product demand seasonality and prediction. Our solution helps in the identification of malpractice by means of finding anomalies in consumption and trip duration.

The Return of the Investment is very fast. Our solution is implemented by connecting their database to our dashboard. They will immediately have access to our services. With such value, worker-hours can be reduced and therefore the small expense on our solution is quickly recovered given the gain in efficiency.

Our solution goes beyond the Business Intelligence paradigm since it is capable of analysing data in-depth given its connection to the state-of-the-art analytic tools, graphs are of the greatest quality and the functionalities can be customised with little addition of code. It is a flexible, agile solution for more than intelligent, wise logistic business.


EOSC-hub helped by providing access to experts and tools, and also guidance and supervision through the process. All improvements with regards to operability of the dashboard are thanks to the guidance and provided support. Through the meetings they also provided a critical environment, where our solution could be analised by other pilots’ managers and presenting our advances there was very helpful.