Cloud-based estimation of marine freight rates
- /
- POC
- /
- Cloud-based estimation of marine...
The logistics chain is crucial for manufacturing, accounting for 20% of final product costs. Currently, tools are slow due to the numerous variables. A software-based solution is developed training a Neural Network to estimate freight costs, using historical data from VGSC and market information.
Start date: 01/11/2015
Duration in months: 18
Problem Description
The logistics chain is an important part of the manufacturing process. Logistics can account for 20% of the final cost of a product for a manufacturing company, and in cases where products need to be shipped by sea, this can rise to 90%. Therefore, finding an optimal transport choice is fundamental.
Goals
New services
Challenges
A software solution must be developed, which takes into account all possible variables in the determination of the optimal transport choice in timeframes compatible with the fast decision-making required by the process.
Innovation results
A software-based solution has been developed to estimate freight costs for specific shipments using VGSC's historical data and market information. The solution uses a Neural Network (NN) and is delivered as a secure SaaS, with performance improving with more data.
Business impact
IMATIA launched a freight cost estimation service, combining data processing, information analysis, and NN knowledge. VGSC tested the SaaS platform, gaining knowledge on freight rate influences from historical data. CESGA developed a HPC Python toolkit for hyper-parametric search on TensorFlow.