Automating and optimizing global requisitions processing at the world's largest shipping company.
CLIENT
- Maersk
SERVICES USED
Data Science
Data Modeling
Artificial Intelligence
For Maersk Procurement, Flowtale were able to dive right into one of our most complex processes, quickly understand our business challenge and translate that into a functional algorithm in around 4 months. The flexible and solution-oriented approach of Flowtale greatly contributed to the success of our project.
- Lars Andersson, Head of Procurement Automation & Analytics, Maersk Line
Challenge
Automation of requisitions processing for a global fleet of shipping vessels by taking all available data into account to reduce costs, optimize delivery port selection and discover new potential suppliers of articles.
Solution
A set of data science algorithms were implemented as a containerized solution. The algorithms are linked together as a single API and integrated with existing business processes. Consistent use of database persistence to track flow and results is used to achieve further AI-based learning.
Automation
The solution uses multiple levels of supply sourcing (catalogue prices, explicit rules and past supply relationships) to automatically infer the best supply solution. The containerized application is available 24/7 as an API, accepting requisitions requests and returning results.
Efficiency
Total requisition costs were minimized by comparing both actual and estimated prices, while explicitly accounting for historical supply statistics.
Focus
The solution algorithmically guides the supplier selection process towards fewer suppliers with completive offerings and item coverage.
Results
A continuously available automation service with a sub-minute round trip batch processing time, achieving the objectives of lower costs, fewer suppliers and the intelligent switching of delivery ports by analyzing outcomes by port in parallel.
100.000+
Requisitions Processed Per Year
213
Worldwide Ports Covered to Date
344
Vessels Served to Date