Ministry of Defence (ACDS Logistic Operations)


MOD (ACDS Logistic Operations) wanted to verify that all logistics lessons had been captured from deployments in Afghanistan between 2002 and 2015. They needed to use the data to ascertain the most important themes in order to more effectively plan for future operations


Working with partners QDA Services, Bricolage extracted data from 22 different reports and documents in PDF, and combined them with 30,120 text fields from a database of defence lessons (extracted as Excel XLS).

This created a corpus of more than 4 million words.

From this we created a bespoke dictionary to extract sentences relating to all lessons learned.  The work was based on an analysis of word frequencies, entity recognition, phrase recognition, link analysis and keywords in context that appeared in the corpus.

We were then able to narrow down the data still further to specifically highlight  lessons related to Logistics by creating a logistics taxonomy with 40 main categories and multiple levels. This was used to extract and classify the lessons relevant  to logistics. These were then analysed for their details, similarities, differences, and reinforcement of topic and classified into broad themes.

Our work enabled ACDS Logistical Operations to be confident that all lessons had been captured from the disparate sources. They were able to summarise the lessons into 8 major logistical themes and use them to gain insight into the complexity of similar issues and in planning for future operations.

For more information please see 'Text Analytics for the MOD'
  • Ministry of Defence (ACDS Logistic Operations)
  • Extraction of technical data
  • Intelligence analysis software
  • Information triage
  • Text mining
  • Extraction of financial data
  • Policing software
  • Swansea University
  • Amnesty International ‘Decoding Darfur’