BIOMETRICS IN A HUMANITARIAN CONTEXT Andrew Hopkins & Justin Hughes 1
Presentation Overview s global footprint The Genesis of Biometrics within The Biometrics Implementation Project Conclusions, the Future and Questions 2
Office of the United Nations High Commissioner for Refugees Established December 14, 1950 (United Nations General Assembly) Mandated to protect refugees and resolve refugee problems worldwide Promotes the right to seek asylum and find safe refuge in another State Seeks options for durable solutions 3
33.9 million persons of concern 4
7,685 staff working globally 5
global exposure 6
Genesis of the Project has attempted with some success to deploy biometric tools globally since around 2003 including fingerprints or iris. On 16 December 2010, announced its policy on biometrics in refugee registration and verification processes. This lead to the strategic decision to develop an integrated solution that can be applied consistently across its network. The policy states that should introduce the collection of biometric data as a regular and routine feature of its registration processes in support of identity verification exercises among refugee populations. 7
The Biometrics Implementation Project In mid-2013, permission was sought from the Biometrics Project Steering Committee to launch a pilot project in a Field Environment This Pilot Phase was aimed at demonstrating that a complex biometric solution could be successfully deployed in environments In addition, the project team sought to understand what factors of the environment would influence the specifics of the technology solution, in particular: Biometric modality Local infrastructure IT infrastructure / Communications 8
Designing the Pilot Phase In order to keep the project momentum, it was deemed important to deliver a Pilot Phase pre-christmas 2013 Selection of a Pilot Phase location: Systems Population size Data control Accessibility Stability Language 9
Finding the Goldilocks Zone Too cold Small Population Just right! 15,000 refugees Too hot Large Population Paper Based Proven Technology New Technology Unstable Population Mobile Population Too Static Population No connection Intermittent Connection Broadband Camp Accessible Camp Urban Too Short 4 weeks Too Long 10
Malawi, Africa 11
Dzaleka Refugee Camp, Malawi 12
Setup & Infrastructure 13
Setup & Infrastructure 14
Setup & Infrastructure 15
Setup & Infrastructure 16
Setup & Infrastructure 17
Setup & Infrastructure 18
Food distribution 19
Food distribution 20
Food distribution 21
Initial Validation 22
Biometric Capture 23
Ration Card Issuance 24
Ration Card Issuance 25
What did we learn? Biometric modality Operator Feedback Population Coverage Refugee Feedback 26
Percentage of Respondents Ease of Use of Biometric Modalities 90% 80% 70% 60% 50% 40% 30% 20% Fingerprints Iris Face 10% 0% Easy Average Difficult Ease of Use Rating 27
Common Issues Fingerprint Capture Iris Capture Facial Capture Cleaning fingers Pressing fingers Repositioning Lowering Demonstrating Opening eyes (wider) Device positioning Demonstrating Sitting / positioning Pose Looking at camera Face wiping 28
Operator Perception Facial Recognition Iris Scanning Preference Speed to enrol Ease of use Fingerprint 100 50 0 29
Fingerprint Capture 3 fingers required re-capture per person on average I always had to recapture little fingers as they were often misplaced on the device first time children were hardest to capture as their hands were often dirty or sweaty 4 times a day, the hardware needed cleaning 30
A reminder 31
Summary Results 16 enrolment stations 4 verification stations Target average of 1,000 per day 16,849 refugees enrolled 314 adjudication cases (91% of which were in the 0-3 age range) 4 week exercise 140 staff 32
Fingerprint Results Ages 4+ 86.7% over four good fingerprints 12% low quality fingerprints 1% no capture Ages 0-3 2% over four good fingerprints 2% low quality fingerprints 96% no capture 33
Fingerprints by age 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% >Four Good Quality Fingerprints Captured 0-3 years 4-10 years 11-18 years 19-50 years 51+ years 34
Iris Results Ages 4+ 98% at least one good iris 1% low quality irises 1% no capture Ages 0-3 14% at least one good irises 0.1% low quality irises 85.9% no capture 35
Irises by age 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Two Good Quality Irises Captured 0-3 years 4-10 years 11-18 years 19-50 years 51+ years 36
Face Results Ages 4+ 79% good face 20% low quality face 1% no capture Ages 0-3 57% good face 42% low quality face 1% no capture 37
Faces by age 90% Good Quality Faces Captured 80% 70% 60% 50% 40% 30% 20% 10% 0% 0-3 years 4-10 years 11-18 years 19-50 years 51+ years 38
Results Population coverage of ages 4+ using just fingers was 86.7% Population coverage of ages 4+ using just irises was 90.39% Population coverage of ages 4+ using fingers and iris was 98.88% 39
Refugee Feedback 40
Connectivity Current systems are country-based Global population data management tool (progres v4) Global biometrics tool 41
Connectivity X and Y connection Two dedicated ISPs serving X connection Highly variable bandwidth 1:N verification times to central database: 25 seconds for iris 35 seconds for fingerprints 42
Conclusions Choice of biometrics Adjudication workload Speed of enrolment and verification System infrastructure What next? 43
It is very often not possible to adapt our infrastructure to fit advanced technology. The technology needs to be adapted to suit the infrastructure we have to work with. Senior Stakeholder 44
Thank you 45
Biographies Andrew Hopkins From Canada, Mr. Hopkins is a Senior Registration Officer with the United Nations High Commissioner for Refugees () based in Geneva. Mr. Hopkins began his refugee work resettling Vietnamese refugees to the United States in 1991 at the US Orderly Departure Program out of Bangkok and Ho Chi Minh City. He has since helped initiate several large-scale group resettlement projects in Kenya, Nepal and Thailand. In later assignments, Mr. Hopkins contributed to the drafting of the Registration Handbook, the development of the first versions and subsequent global rollout of progres, s registration and case management software. He has experience working and supporting population data management in emergency operations most recently supporting the Syria situation in the Middle East. Mr. Hopkins is currently leading s global biometrics business effort to ensure that refugee identities are established and maintained throughout the refugee lifecycle. Justin Hughes From the United Kingdom, Mr. Hughes is a consultant with PA Consulting, a London-based management consultancy. Prior to joining PA Consulting, Mr. Hughes was employed by De La Rue and played a key role in the delivery of the new UK biometric passport, as well as many other identity management projects around the world. His last role was as Head of Procurement and Supply Chain for the Identity Systems division of De La Rue where he oversaw all sourcing, supply management, risk analysis and innovation support for identity management projects. Since joining PA Consulting, Mr. Hughes has spent much of his time supporting with the sourcing, development and delivery of the global biometrics system as well as leading PA Consulting s work with the United Nations in Geneva, Copenhagen and New York. 46