• Cranfield University partners with UN to automate painstaking research for opium monitoring programme
  • Globally, over 80% of illicit opium is produced in Afghanistan

Researchers from Cranfield University are using their world-leading research to help the United Nations (UN) develop artificial intelligence (AI) that supports the monitoring of illicit opium production in Afghanistan.

An Implementing Partner Agreement between Cranfield and the United Nations Office on Drugs and Crime (UNODC) will see academics use AI to interpret data from satellites to track where the crops, used for drug production, are grown in Afghanistan.

Globally, over 80 per cent of illicit opium is produced in Afghanistan, and the data provided by Cranfield will help the UNODC work with the country’s government to monitor the extent and evolution of these crops. The survey data enables UNODC to help governments decide how to tackle the issue of opium production.

Dr Daniel Simms, a Lecturer in Remote Sensing at Cranfield University, said that the university’s research on illicit poppy cultivation in Afghanistan would form an essential part of the opium monitoring programme, and that work to automate the process using AI would save many hours.

“This is a hugely significant project to be involved with. In terms of Afghanistan, we’re talking about opium production on a grand scale – something like the combined size of 500,000 rugby pitches – so a huge amount of cultivation.

“The UN recognises our world-leading expertise in the technological development of AI to assess crop cultivation, and we are really pleased to be able to deliver this vital work for them.”

The partnership with the UN lasts until July 2023.

You can watch an earlier TedxOpenUniversity talk from Dr Simms on YouTube.

For more information on the project please visit our website or read more about the project here.

Cranfield University’s agrifood MSc courses explore the integrated nature of food supply chains and the ongoing need to increase their economic and environmental sustainability. You can find out more on our Environment and Agrifood web page.

Related publication

Simms DM, Hamer AM, Zeiler I, Vita L & Waine TW (2023) Mapping agricultural land in Afghanistan’s opium provinces using a generalised deep learning model and medium resolution satellite imagery, Remote Sensing, 15 (19) Article No. 4714.