Build a stock management interface in order to predict and optimize on site capacity and control the costs generated by the warehouse



DAHER is an aeronautic actor with five main activities: 

  • Aeronautic constructor with TBM series
  • Aerostructure and systems (Belly fairing, trapdoor…)
  • Logistics
  • Nuclear services
  • Valve constructor

Daher employs around 8500 co-workers and its revenue in 2016 was over 1 billion euros. 

The challenge concerns a site in Saint-Nazaire where DAHER assembles belly fairings for Airbus.

 Baily Fairing: Zone 2

Baily Fairing: Zone 2

The activity on site is divided in several parts:

  • Reception of all primary parts
  • Storage
  • Kitting
  • Shipping to Airbus Pre-FAL and FALs (FAL : Final Assembly Line)

In its warehouse in Saint Nazaire, Daher collects a lot of references from different suppliers. To assemble a Belly fairing, the activity is divided in subtasks that are called work packages. 5 work packages are needed for a complete Belly Fairing. The production is around 480 aircrafts per year with over 4 declinations.

A well-controlled storage of pieces can generate an important cost saving for Daher. It is estimated that a better stock management will generate a saving of more than 10% on the stock cost. Ultimately, stock optimization will improve Daher’s competitiveness.




Based on Airbus pace of order and other critical parameters, DAHER would like a predictive algorithm to better anticipate and optimize its stock under constraints. The algorithm should enable on-site logistics coordinators to predict their target storage.


Predix helps you develop, deploy, and operate industrial apps at the edge and in the cloud. The platform helps you securely connect machines, data, and analytics to improve operational efficiency.

Success indicators

The success will be measured with:

  • An accurate algorithm able to give the optimal stock for each work package and elementary part
  • Compared to an old belly fairing order, the same level of quality with less pieces in stock
  • One differentiating criterion to assess solutions would be the ability of the algorithm to be generalized to other Daher activities and industrial sites.


Data is collected from classic ERPs. There is real time access and data is property of DAHER. There is a historic of more than 10 years of available data.

There are a plenty of input data variables :

  • Order from Airbus
  • Order from DAHER to suppliers
  • Time of delivery
  • Stock IN, stock OUT
  • Need in pieces for each work package

business scenario

Daher has a real interest in stock optimization. If the app is a success, other projects could be deployed in partnership on other sites in France and abroad in Mexico, USA, China.


Capture d’écran 2017-09-29 à 16.58.14.png