Realizing Digital Identity in Government: Prioritizing design and

implementation objectives for Aadhaar in India

Umar Bashir Mir

Ph. D., Student
Indian Institute of Technology Delhi, New Delhi
National Capital Territory of Delhi, India
Email: [email protected]

Arpan K. Kar
Associate Professor
Indian Institute of Technology Delhi, New Delhi
National Capital Territory of Delhi, India
Email: [email protected]

Yogesh K. Dwivedi
Professor
Swansea University, UK
Email: [email protected]

M. P. Gupta
Professor
Indian Institute of Technology Delhi, New Delhi
National Capital Territory of Delhi, India
Email: [email protected]

R.S Sharma
Chairman
Telecom Regulatory Authority of India (TRAI), New Delhi
National Capital Territory of Delhi, India
Email: [email protected]

Abstract: With the increasing levels of digital transformation, focus on digital

identities of individuals is increasingly getting prominence. It is the information

captured as part of the identity surrounding the citizens which decides what

services and products one is entitled to and can access. At present, there are still
around 1.1 billion people in the world without any official identity. To address

this concern, United Nations through its 16th Sustainable Development Goals

(SDGs) recommended governments to provide their citizens with unique

identities by 2030. India’s Aadhaar is one significant step in this direction as it

has already reached over 1 billion enrolments in India. However establishing a

national digital identity program successfully requires expertise, time, and huge

financial commitments. This paper takes Aadhaar as a case study and uses Design

Theory (DT) and Critical Success Factor theory (CSF) as a theoretical lens and

attempts to evaluate design and execution choices made during the tenure of the

project. The study also identifies and prioritizes primary goals of Aadhaar based

on the secondary data, expert opinion through a Focus Group Discussion (FGD)

and subsequently systematic prioritization using mixed research methodologies.

The expert opinion from the FGD was analyzed using the Best-Worst method

(BWM), followed by the Total Interpretive Structural Modeling (TISM) method

and Matrix of Cross Impact Multiplications Applied to Classification (MICMAC)

analysis. The study identified uniqueness, security and privacy as the top priority

goals in an identity system and is 11 times more crucial than scalability which is

identified to be of lesser priority. These findings from this study could be

considered as a reference for other countries that aim to develop and implement

digital identity for its citizens.

Keywords: Digital Identity, Aadhaar, Design Theory, e-Governance, Critical

Success Factor, Best-Worst, TISM

 1. Introduction

Identity is a multi-faceted concept, and defines an individual uniquely. It is comprised of

those attributes that makes an entity unique and distinguishable from others (Olson, 2015).

(Ben Ayed, 2014) also defined identity as a set of qualities and characteristics, by which an

entity can be defined, distinguished and recognized in comparison to other entities. With a

focus on context, International Telecommunication Union (ITU) has defined identity as a set

of one or more attributes that distinguish a particular entity within a context (ITU-T, 2010).

The International Organization for Standardization (ISO) defines digital identity as “item

inside or outside an information and communication system, such as a person, an

organization, a device, a subsystem, or a group of such items that has a recognizably distinct

existence.” (ISO, 2011). Further, World Economic Forum (WEF) defined digital identity as

“collection of individual attributes that describe an entity and determine the transactions in

which that entity can participate” (Mcwaters, 2016).

India became the first nation with a population exceeding 1 billion citizens to

implement digital identity for all its citizens. With a population of 1.3 billion (Bank, 2016),

India is second most populous country in the world. For such a country to roll out a program

like Aadhaar, which is world’s largest digital identity program as of July 2019, by Unique

Identification Authority of India (UIDAI) in itself is a big challenge considering various

practical challenges like high illiteracy rate (26% as per 2011 census), diverse cultures,

political beliefs, varied demographics and many more. The primary purpose of Aadhaar –

Unique Identification and Online Authentication, is achieved by the technological

advancements in biometrics (Gelb & Clark, 2013). Aadhaar is a combination of 12-digit

random number (Barnwal, 2015) tightly coupled with biometric details of an individual like a

photograph, fingerprints, iris and demographic information like age, date of birth, gender and

address. India has issued Aadhaar cards to more than 1 billion residents after initial launch
(Dixon, 2017). As of February 2019, more than 1.2 billion people have already been issued

Aadhaar numbers (UIDAI, 2019) and hence it surpasses the US's biometric project called

VISIT, which was earlier considered to be the world's largest biometric project (Epstein,

2008). This current study is much needed for giving a roadmap for implementation of digital

identity considering the recent recommendations by United Nations in its 16th SDG that each

individual should be provided a legal identity by 2030 (UN, 2016). One of the driving

motivation for this study is that at a global level, despite the focus brought out in the 16th

SDG of the UN, 24% of the developing countries do not have any form of digital identity

system, and among others only 3% of the countries have foundational identity system that

could be used in both online and offline domains (BankWorld, 2016; Segovia, Álvaro, &

Enríquez, 2018). Countries that do not possess any national identity scheme could use this

roadmap such that the desired objectives of identity for all is achieved in a timely and cost-

efficient manner. For meeting such an objective, this study is useful for benchmarking design

and execution objectives of digital identity programs, considering its developmental cost,

reach and the technologies used.

E-government projects fail because of the multiple reasons (Aladwani, 2016;

Anthopoulos, Reddick, Giannakidou, & Mavridis, 2016; Elkadi, 2013). One of the main

reasons for the failure of e-government projects is the divergent interests and differences in

the stakeholder expectations from the project (Sivamalai, 2013). Further, Sivamalai using

Social Construction of Technology (SCOT) framework studied how different stakeholders

perceive Aadhaar differently and how the analysis of design decisions before implementation

could solve this problem. The present study takes it forward from there and evaluates the

design and implementation decisions made during the tenure of the project such that the risks

of the possible project failure could be mitigated. This study makes an attempt to breakdown

the complexity of developing a digital identity system like Aadhaar in a way that it becomes
easier for others to follow the process and develop a biometric based digital identity system

based on their requirements.

To understand the implementation of Aadhaar, an in-depth analysis of Aadhaar from

planning to implementation phase is required. The selection of over-arching goals for

Aadhaar and multiple criteria associated with each goal can be considered as a MCDM

problem because each goal is dependent on multiple factors/criteria that need to be taken care

of right from the planning phase. From theoretical perspective this study uses CSF and DT to

study overarching goals and their design and executions choices respectively. Both the

theories have been used extensively in the e-government literature e.g. (Rana, Dwivedi, &

Williams, 2013) used CSF to identify challenges and barriers of e-government adoption,

(Akhtar Shareef, Kumar, Kumar, & Dwivedi, 2014) identified ability to use and assurance to

use as the critical factors in the adoption of electronic government, (Shah, Braganza, &

Morabito, 2007) identified organizational factors critical to e-business, and (Bergeron &

Bégin, 1989) highlighted the use of CSF in evaluating information systems. Similarly, extant

research literature about DT is nicely covered (Agogué & Kazakçi, 2014). Studies have

proposed DT for market surveillance (Li, Sun, Chen, Fung, & Wang, 2015), digital platforms

supporting online communities (Spagnoletti, Paolo, Resca, & Lee, 2015), and for developing

policy alternatives (Pluchinotta, Kazakçi, Giordano, & Tsoukiàs, 2019). Some of general

utility areas of DT are design and development of policies (Esfahlan & Valilai, 2019;

Howlett, 2014; May, P.J, 2003; Mintrom & Luetjens, 2016) and decision making (Le

Masson, Hatchuel, Le Glatin, & Weil, 2019).

This study attempts to address following research questions:

• What are the overarching design and implementation goals of India’s digital

identity – Aadhaar?
 • How can we establish the priority among these overarching design and

implementation goals?

• What are the influencing factors that made it possible to design, develop and

implement such a large scale biometric digital identity program?

To answer these questions, a MCDM method, namely the Best-Worst MCDM method

has been adopted for evaluating the experts’ feedback gathered from a multi-stakeholder

workshop. All these experts were identified from the government and private sector having

direct and senior role in the implementation of Aadhaar. Inputs from experts are analyzed and

transformed into the list of weighted criteria based on their significance corresponding to the

primary goals of Aadhaar. This prioritization will be helpful in forming an implementation

and risk assessment roadmap for large projects similar to Aadhaar and can enhance the

probability of a project to be successful. It can also increase the transparency, acceptance and

utility of a particular project. Further, two well established methodologies, TISM and

MICMAC analysis are used to verify the BWM results.

The remaining sections are organized as follows: Section 2 describes the unique

identification concept primarily from biometrics perspective, including its functional utility,

technical architecture and process flow. Section 3 presents Aadhaar as a digital identity case

study. Section 4 discusses about the research gaps in the domain and contribution made. The

theoretical lens for this study is covered in section 5. Section 6 introduces the focus group

followed by data analysis part of the study using BWM for prioritization. Verification of

BWM output using TISM and MICMAC is covered in section 7 and finally, discussions

about the theoretical and practical contribution of this paper along with the future research

directions are explained in section 8.

 2. Literature Review

This section is subdivided into three sub-sections. The first subsection focuses on the utility

of biometrics for citizens from a functional perspective. The second sub-section explores the

technical architecture for such a solution and process mechanism for biometrics in practice is

explained in the third sub-section.

2.1 Biometrics for Citizens – The Functional Utility

Biometrics could be used for identity recognition based on various biological traits like voice,

iris, face, fingerprints, palm, DNA, ear, retina or behavioral characteristics like handwriting,

signature and body movements, also called gait (Hoang & Caudill, 2012) Among all;

fingerprints are the oldest biometrics in use. Traditionally, fingerprints were used in ink and

paper documents for legal purposes. In the recent past, DNA as biometric has also gained

much interest among researchers. However, any biological or behavioral trait can be used as

biometric characteristics as long as it agrees to the criteria shown in Table 1.

Table 1: Biometric selection criteria

Criteria Description
Universal Every individual should have the particular biometric attribute. The attribute
must be scarcely lost to accidents or health related illness.
Unique Biometric attribute should be different for each individual. Attribute should
possess distinct properties such that one individual could be distinguished from
other.
Permanent Attribute should remain unchanged indefinitely. It should be constant over the
period of time and should not be subjected to any major change.
Recordable Once captured, biometric attribute should be storable such that it becomes easy
to handle and perform operations on it.
Based on the unique features of biometrics, biometrics based technologies have a wide

scope of applications like logical or physical security (Hodeghatta & Nayak, 2014),

surveillance (Bouchrika, 2017), healthcare (Marohn, 2006), law enforcement (A K Jain,

Ross, & Prabhakar, 2004), time and attendance (Fenu, Marras, & Boratto, 2018), and

electronic signatures (Nunno, 2000). These studies highlighted that the design and execution
goals of any digital identity should address universality, distinctiveness, permanence,

collectability, performance, acceptability and circumvention.

2.2 Biometrics for Citizens – The Technical Architecture

It is found that researchers started publishing about biometric systems from 1960's. In the

year 1963, Mitchell Trauring published his work on fingerprint matching on automated

biometric recognition (Trauring, 1963). After that, other biometrics have been used in various

automated biometric systems like voice (Pruzansky, 1963), signature and Hand geometry

(Mauceri A.J, 1965) and iris (Daugman, 1993) systems were also developed subsequently.

Biometric systems have evolved over a period and have improved considerably

regarding performance, accuracy, and usability. Any new biometric technology is evaluated

against the benchmark set by National Institute of Standards and Technology (NIST) before it

can be made available for commercialization. NIST evaluation process is very complex and

takes into consideration various test conditions before finalizing the results (NIST, 2013).

The detailed explanation of these test conditions adopted by NIST is beyond the scope of this

paper. However, a few important parameters that Digital Identity Systems (DIS) use globally

has been illustrated in appendix section A.1. Some of the state-of-the-art biometric

technologies can achieve greater accuracy (Phillips, Flynn, & Bowyer, 2017). An overview of

the comparison of various biometric identifiers based on their application and design goals is

illustrated in Table 2.

Table 2: Important factors in biometric identity of citizens, extended from A. K. Jain et al.,
(2004)
Hand geometry
Thermogram
Finger-print

Biometric
Hand vein

Keystroke

Palmprint

Signature

Identifier
Retina
Facial

Voice
DNA

Odor
Face

Gait
Ear

Iris

Universality H M H H M M M M H L H M H L M
Distinctiveness H M L H H L M M H L H H H L L
 Permanence H H M L H L M M H L H H M L L
Collectability L M H H M H H M M M L M L H M
Performance H M L M H L M M H L L H H L L
Acceptability L H H H M H M M L M M M L H H
Circumvention L M H L M M M L L M L M L H H
Inbuilt security H M M H M M M H H L L M H L M
Processing cost H M L M L H M H H L M M H L L
Convenience L M H M H L M L L H M M L H H
Robustness H M H M L L L H H L M M H L L
Utility L L H M H M M L H M L M H H M
Where H = High, M = Medium, and L = Low labels are based on the authors perception.

2.3 Biometrics for Citizens – Process Flow

The National Science and Technology Council provides the following overview of biometric

system components: “A typical biometric system comprises of five integrated components: A

sensor is used to collect the data and convert the information to a digital format. Signal

processing algorithms perform quality control activities and develop the biometric template.

A data storage component keeps information with which new biometric templates will be

compared to. A matching algorithm compares the new biometric template to one or more

templates maintained in data storage. Finally, a decision process (either automated or human-

assisted) uses the results from the matching component to make a system level decision.

Fig. 1. Process blueprint for the verification process through a biometric system

Fig 1 illustrates a process flow of a typical biometric system and categorizes processes

into three stages, citizen enrolment, citizen on-boarding and decision making. Citizen

enrolment is the first stage where in data is collected from the citizens and made ready for

further processing by cleaning and filtering data from noise. In the second stage biometric
templates are generated after feature extraction and in the final stage these templates are

permanently stored in the database for decision making.

Classification of authentication systems can be done based on what an individual has

(e.g., tokens, smart cards), what the individual knows (e.g., PIN, passwords) and what the

individual comprises of (e.g., iris, fingerprints). As per Census 2011 there are about 104

million elderly people aged above 60 years who often forget access credentials and is

expected to grow to 173 million by 2026 (GOI, 2016). Also such users often share their

credentials to others to access the services, due to which two factor authentication systems

became popular (Singh, Cabraal, Demosthenous, Astbrink, & Furlong, 2007). Similarly,

biometric systems could also get compromised, as demonstrated by Professor Tsutomu

Matsumoto at Electric Imaging 2002 conference. He developed a fake finger with gelatin and

used a simple digital camera and a computer to fool a biometric device 80% of the time (BBC

News, 2002). However, with recent advancements in biometrics technology can detect fakes

by considering factors like sweat pores, conduction properties and finger on contorts the

surface of a biometric scanner (Anil & Sharathchandra, 2001).

It is important to note that biometric-based systems are not entirely foolproof (Pagnin &

Mitrokotsa, 2017). Even though biometric technology has come a long way in the last five

decades, it still has challenges and issues that are not sufficiently addressed yet (Bálint &

Bucko, 2013; Chandra & Calderon, 2005; Anil K. Jain, Nandakumar, & Ross, 2016; Uludag,

Pankanti, Prabhakar, & Jain, 2004). Resolving biometric technology issues in it-self is a

separate research area, and researchers are actively working in this field (Abate, Marcialis,

Poh, & Sansone, 2019; Arutyunov & Natkin, 2010; Baichoo et al., 2018; P.Down & J.Sands,

2004).
3. Aadhaar – The Indian Case Study

In 2009, Government of India came up with a proposal to provide every citizen of India a

Unique Identification which can be used to provide benefits of various government schemes

to desirable citizen directly. Considering the fact that India is the second most populous

country in the world it was extremely vital to use some technology which can effectively and

efficiently serve the purpose. With continuous improvement in accuracy and reliability of

biometric technology, it was one of the best options available for identifying people uniquely.

The primary purpose for rolling out Aadhaar was (a) to ensure proper utilization of

government subsidies (b) to provide a unique identity to every citizen of India which can be

accepted as identity and address proof throughout India (c) to tackle illegal immigrants (Dass,

2011; Ronald, Elizabeth, Noopur, & Neil, 2017; Zelazny, 2012). Technological architecture

of UIDAI is shown in Fig 2. For the implementation of Aadhaar, it was very vital to have a

mechanism that can deal with duplicates, be time efficient, scalable and should be feasible to

integrate with other existing systems like Public Distribution Systems (PDS). The

government took a transformational decision to implement AADHAAR which is based on the

demographic and biometric details of a person. To decide which biometrics to use, a

Biometric Committee was constituted which presented its report (Zelazny, 2012) to UIDAI.

Finally, demographic details, photograph, fingerprints of 10 fingers, and iris of both eyes

were included as necessary data inputs for the issuance of unique Aadhaar number (UIDAI,

2009). Rolling out an identity system which could suffice such a heterogeneous population

was a big challenge for UIDAI. Aadhaar was launched with the intention to provide legal

identity to the residents of the country so that they could avail various welfare benefits which

they were denied earlier because of the lack of official identity documents. In addition to that,

it was also intended to reduce corruption, reduce intermediation and agency costs, avoid
identity related frauds and most importantly to increase participation of people in various

government sponsored welfare schemes.

It can be observed from the Table 2 that no single biometric technique can out-perform

others in all factors and hence a combination of biometric technologies is required based on

the importance of each factor in a particular application. As a pre-study to establish the

feasibility of Aadhaar, we performed a first-hand check of the methods of digital identity, as

illustrated in Table 2. Our exploratory study indicated that biometrics used in Aadhaar -

Photograph, Finger-print, and Iris, complement each other in all factors. It indicates that

Aadhaar has required biometrics in place to deal with de-duplication, scalability, and

uniqueness of its Central ID Repository.

Fig. 2. Technology Architecture of Aadhaar (Adapted from UIDAI, 2010)

Where, FK = Foreign Key, PK = Primary Key, PDS = Public Distribution System, POS = Point of
Sale, RDPR = Rural Development and Panchayath Raj, and PAN = Permanent Account Number
 India is the only country with such a large scale for national identification of its citizens

(exceeding 1 billion) through Aadhaar. As of February 2019, more than 1.2 billion citizens

out of 1.3 billion have enrolled under Aadhaar (UIDAI, 2019). Notably, India is not the only

country to implement national identification program, some of the other countries as shown

in Table A.1 have similar identification programs like Aadhaar although their purpose varies.

4. Research objectives

The utility, design and governance of digital identity system is an emerging area of research

which has not been studied much (McKinsey, 2019). Development, governance and

implementation of e-ID for public e-service delivery is quite challenging and requires a lot

coordination (Melin, Axelsson, & Söderström, 2016). Some Journals have published special

issues specifically for electronic identity and eGovernment related research covering the

multi-dimensional perspectives of e-IDs and its challenges and applications (Gal & Whitley,

2011; Irani et al., 2016; Meier & Terán, 2019; Whitley, Gal, & Kjaergaard, 2014). In

(Lentner & Parycek, 2016), authors have examined different legislative approaches adopted

in four European countries with focus on identification and authentication in customer-to-

government eGovernment services, similarly, (Seltsikas & O’Keefe, 2010) studied challenges

faced by stakeholder in developing, implementing and maintaining trusted e-IDs, (Hedström,

Karlsson, & Söderström, 2016) identified usability, attitude, behavior and privacy concerns

as primary challenges in implementing eID in healthcare.

To the best of our knowledge, there is no such study found in the literature which has tried to

prioritize the attributes of any biometric identity project like Aadhaar and their

interdependence with each goal. Majority of the research on Aadhaar explores the application

side of it, e.g., electronic voting, payment systems, continuous verification of online exams,

financial inclusion, healthcare sector and in government welfare schemes. Therefore, the
motivation for this research is to analyze Aadhaar project from design and implementation

perspective thoroughly. This study deals with three main objectives:

• To identify overarching goals of Aadhaar and their significance in a biometric based

digital identity system.

• To identify design and execution goals of Aadhaar corresponding to each overarching

goal and their significance and inter-linkages with respect to goals.

• To establish how these goals are linked to each other in meeting the overarching

objectives of the project.

This study contributes to three perspectives. First, it prioritizes the attributes of Aadhaar

for each goal based on the expert opinion. It gives us an opportunity to uncover how Aadhaar

project was approached during its incremental implementation phase which led to its

successful implementation. This study could act as a reference for implementation of similar

large size biometric programs in future.

Second, methodologically, we also attempt to demonstrate the application of the BWM

method on a real world, highly complex group decision-making problem. This study could be

used as a valuable case for the validation of BWM on real-world decision problems involving

more than a single decision maker. Further, the results of BWM is again verified by TISM

and MICMAC methodologies for identifying consistency in results. The case study highlights

how such an approach can be effectively used for prioritization and decision making, for

complex problems involving multiple factors.

Thirdly, from theoretical perspective, DT for digital identity system is integrated with

CSF for a better understanding of digital identity systems. Both these theories are well

explained in the literature as mentioned in the introduction section and in the section for

theory development.
5. Theoretical lens

Theories like Deferred Action Theory (DAT), systems theory, and stakeholder theory are

options available that could have been used in this study, but DT is chosen over others

because DT is more suitable for this study as elaborated subsequently.

DAT is relatively new as compared to DT. The basic building blocks of DAT are

planned action, emergence, and deferred action (Schneberger, 2012). It helps to discover the

impact of emergence on the company and system design (Patel, 2007). DAT is suitable for

conceptualizing systems that are operating in a dynamic neighborhood and thereby causing

systems and organizations to be emergent (Patel & Hackney, 2010). DAT is relevant and

could have been used in this study as the primary theoretical lens but DT is preferred over

DAT because the researchers across specializations have widely accepted the former one

when stable systems are planned. Moreover, this study analyses an existing systems rather

than conceptualizing a new non-existent system and the possibility to evaluate a product and

its development process followed makes DT a better choice.

In similar lines, systems theory is used to explore a complex set of interacting

elements in a system (Daniël F M Strauss, 2002; Von Bertalanffy, 1956). Systems theory and

its variants have been widely applied in various research areas like personality development

(Millová & Blatný, 2015), motivational development (M.E. Schneider, 2001; Marianne E.

Schneider, 2015), organizational behaviour management (Ludwig, 2015), healthcare

(Champion, Kuziemsky, Affleck, & Alvarez, 2019; Clacy, Goode, Sharman, Lovell, &

Salmon, 2019), governance (Timoshenko, Kuruppu, Badshah, & Ambalangodage, 2020;

Zahra Mansoor & Williams, 2018), and marketing (Mele, Pels, & Polese, 2010). Systems

theory does not provide a sophisticated mechanism to identify the problems that designers

may encounter (Buchanan, 2019). The theory is based on the assumption that a system is

composed of the entities that interact with one another, which in turn depicts a system as a
self-regulating entity. The focus of this study is to analyze the development and

implementation of Aadhaar which is not the main focus of systems theory and this does not

make systems theory a best choice for this study as compared to DT.

Further, Stakeholder Theory (ST) is a proven and widely applied theory in various

areas of research for focusing on groups which are vital to the survival of the organization

(Freeman, 1999; Stieb, 2009; Schneberger, 2012; Singh, Kar, & Ilavarsana, 2017). The

significance of stakeholders in e-governance research has been extensively acknowledged in

the literature (Knol, Janssen, & Sol, 2015; Brooks, Janssen, & Papazafeiropoulou, 2018;

Janssen & Estevez, 2013; Singh et al., 2017). ST is relevant in the context of this study and

could be used as a central theoretical lens, but because the main focus of this study is to

analyze the design and implementation process of Aadhaar explicitly, we have strictly

restricted our analysis to single stakeholder category that is primarily responsible for the

development and implementation of Aadhaar. While extending this study, we will be taking

multiple stakeholders into consideration like citizens, NGOs, and other public and private

sector participants.

This study attempts to validate the Aadhaar biometric identity system through the lens of DT.

DT is often applied to both processes and/or products (Gregor & Jones, 2007). In this

research we have used the lens of information systems DT (Walls, Widmeyer, & El Sawy,

1992) as a baseline for analyzing the project of Aadhaar system.

5.1 Design Theory

DT is a widely accepted and implemented theory (Hatchuel et al., 2016). DT focuses on the

importance of early stakeholder engagement (Liedtka, King, & Bennett, 2013). It evaluates

and examines design as a concept and enables us to verify if the product and the process

followed to develop this product, satisfies the fundamentals of the DT.

DT has been defined through multiple lenses like prescriptive by (Walls et al., 1992),

practical by (Goldkuhl, 2004), principle based by (Markus, Majchrzak, & Gasser, 2002)

,basis for action by (Gregor & Jones, 2007) and dualist construct by (Simon, 1996; Walls et

al., 1992). Walls divided DT for information systems into two major components “Design

Product” and “Design Process” and defined DT for information systems as “a prescriptive

theory based on theoretical underpinnings which says how a design process can be carried out

in a way which is both effective and feasible”.

In this study we have applied DT to analyze the process of developing a digital identity

system in India. Purpose of understanding and breaking down the whole developmental

process of Aadhaar using the DT is to validate the approach adopted which in turn will be

beneficial for the countries that might consider Aadhaar as a reference to build their own

similar biometric identity system. We use DT to provide an explanatory view of an already

implemented digital identity system –Aadhaar. Theoretical approach stresses upon “why” and

“how” aspects of the system components. Meta-requirements justifies why a particular goal is

important and meta-design explains how to achieve a particular goal with the help of design

and execution choices. Components of digital identity DT using CSFs are shown in Fig 4.

5.2 Identification of CSFs of digital identity system

Identification of vital parameters in Aadhaar system is studied through CSF theory approach.

Data was collected from the existing academic research articles, official reports from UIDAI,

news articles, and from experts who were directly involved with the Aadhaar project right

from the beginning. Focus was to cover all the preliminary goals of Aadhaar no matter how

latent they are (refer Table A.2 in appendix for details).

Critical success factors theory: Our research is rightly placed within the theory of CSF.

CSF theory is defined as “the limited number of areas in which results if they are satisfactory

will ensure successful competitive performance for the organization”. CSFs were originally
defined by (Boynton & Zmud, 1984) as “those few things that must go well to ensure

success”. For any large project, the most tedious and challenging tasks are to take right

decisions at right time, without taking any risk for granted. The implementation of CSF

theory enables to facilitate stakeholders to focus on the significant factors that could lead to

achieve a desired goal successfully (Bai & Sarkis, 2013). Thus any initiative taken by an

organization must ensure that the performance of critical factors remains high or else there is

a possibility that the target goals of an organization may not be achieved.

We conducted a comprehensive literature review of various secondary data and research

papers related to digital identity across economies from the Scopus database. Apart from

research articles, we also considered official reports from government especially UIDAI and

news articles published by some of the leading online news portals. A total of forty

preliminary goals were identified that are refined regrouped, renamed and classified into

fourteen generic themes out of which nine were selected as CSFs (these form the Meta-

requirements) by experts (see Table A.2). Initial classification was done by the authors based

on their experience and understanding of literature and classified forty preliminary goals into

fourteen themes. Out of fourteen themes, nine were selected by experts after building

consensus for the same in three iterations. The final nine accepted labels are shown in Table

A.2 in appendix section, and are used for further analysis and address our first research

objective i.e. to identify CSFs of Aadhaar. The overall research roadmap is shown in Fig. 3

below.
 Fig. 3. Research roadmap to identify CSFs of Aadhaar

5.3 Meta-Requirements of digital identity system derived from CSFs

It identifies the list of goals on which theory is applied. For the identification of goals

literature review is done that resulted into an extensive list of objectives which was then

refined i.e. duplicates were removed, similar ones were combined, and loosely related ones

were dropped from the final list based on expert opinion. From the initial forty preliminary

objectives, nine goals were shortlisted using CSF theory. These nine goals form meta-

requirements for this study.

5.4 Design Method of digital identity system

Design method in this case is a composition of two sub-processes, identification of design

choices and execution choices. It was observed that some of the design choices are related to

multiple meta-requirements but vary in degree of significance. Dependence of design and

execution choices on each meta-requirement is shown in Table A.8 in appendix section. The

design and execution choices focus on three aspects of implementation i.e., operational part

focusing on the “how” aspect of fulfilling meta-requirements, technological part focusing on

the set of technologies needed in the system and social aspect focusing on the societal

parameters that play a vital role in the acceptance and effectiveness of the system.

Design choices: Corresponding to nine meta-requirements identified, a total of 17

design choices were made during the tenure of Aadhaar system. The set of design choices

were identified from the extensive literature review and verified by the experts before

processing it further in the study. Design choices ensure that corresponding to each meta-

requirement, a set foundational design criteria are fulfilled.

Execution Choices: Corresponding to 9 meta-requirements identified, a total of 23

execution choices were identified in a similar manner as meta-requirements. It is observed

that some of the choices are common in design and execution choices of meta-design (see

Table A.8 in appendix). The relation between design and execution choices is many-to-many.
 Fig. 4. Components of Digital Identity DT Using CSFs

5.5 Meta-validation of system

Aadhaar digital identity system is tested in two phases. Based on the meta-requirements and

meta-design, UIDAI in 2010 conducted a proof-of-concept study in the three states of India

i.e. Bihar, Karnataka, and Andhra Pradesh. This helped in identifying the lacunas in the initial

version of the scheme. The gaps identified in critical areas like workflow, policy, and system

design were addressed in the modified version of scheme and then launched in other parts of

the country with corrective measures in place (Zelazny, 2012). Prototyping is considered as a

good low-cost tool to gain design information in the early part of design process (HALL,

2001). Another signal depicting successful implementation of Aadhaar could be that more

than 20 countries are keen to implement Aadhaar and its underlining technology for its

citizens (OECD, 2018). We tried to summarize impact of Aadhaar corresponding to its goals

and understand if intended goals of Aadhaar are being achieved in practice or not. Evidences
suggest that goals are being achieved to a larger extent but there are some cases were

progress is still slow. List of evidences corresponding to each goal are shown in Table 3 and

provides a birds-eye-view about the accomplishments of Aadhaar. It tries to explain the

impact of each goal in practice.

Table 3: Validation of Aadhaar

Goal Validation check

Building it as a Linking 252 welfare schemes; distributed compute platforms; serving
platform billions of requests daily; data sharded across multiple databases
Future-proofing of Use of open-source technology; open APIs for linking services;
technology Aadhaar number is blocked on the death of the holder;
Data Security and Option to generate Virtual ID via Aadhaar; limited access to stored
privacy personal information; No option to download data; 2048 bit
encryption with 256 bit hash
Scalability Use of open standards; supports large scaling of enrolments and
authentications;
Inclusion 99% Coverage; 0.14% failure to enroll; started face ID for enrollment
Uniqueness of IDS Random number; necessary inclusion of Iris demonstrated with proof-
of-concept
Cost Optimisation less than 100 Indian Rupees per person
Speed Time required to enroll a single person is approximately 3 minutes; 1
to 4 million enrolments a day; demand based allocation of manpower
for enrollment
Resident 36000 enrolment station operated by 83 agencies via 400 registrars
Convenience across 32 states and union territories; started face ID for enrollment;
home delivery of Aadhaar via speed post in less than 30 days;
Another approach used for meta-validation of the system is by analyzing the research

literature on Aadhaar. We searched for “Aadhaar” keyword on Scopus which is the largest

peer-reviewed database. Only journal articles were taken into consideration and a total of 76

articles were retrieved out of which 38 articles were found relevant for the study. Authors

classified 38 articles manually into three categories based on whether article is in support (17

studies), against (15 studies) or neutral (6 studies) towards the Aadhaar project.

6. Research Methodology

This study adopted focus group methodology to identify the confidence levels between the

over-all objectives, design, and execution of Aadhaar project as it is cost effective (Morgan,

1996) and time efficient (Caroline Tynan & Drayton, 1988) in case of in-depth information
retrieval about a particular topic is needed. FGD is very useful when there is not much

research literature available on a particular topic (Krueger & Casey, 1994), which holds true

for this study. Focus group methodology also known as group interviews is a well-known

interactive and systematic technique for receiving the opinion of experts on particular issues.

It has been used in the various domains like supply chain management (Lambert & Enz,

2017), smart city selection (Kumar, Singh, & Gupta, 2018), business communication

(Hartman, 2004), information systems (Burgess, 2010), and logistics (Coule, 2013;

O.Nyumba, Wilson, Derrick, & Mukherjee, 2018). FGD is an amalgamation of people from

similar backgrounds or experiences together to discuss a particular topic. It falls under the

qualitative research category where questions are asked based on the perception, beliefs,

opinions or ideas. All members of FGD are free to talk with each other and have discussion

based communication. Typically the number of people in a group varies from 6 to 12 people

(Wilkinson, 1998). This whole discussion is led by an interviewer specifically called

moderator who is responsible for directing the overall discussion.

In this study, an eight-member focus group was formed, each related to Aadhaar project

directly and having more than fifteen years of experience in developing and implementing

eGovernment projects. The eight experts ranging from senior management professionals,

Indian Administrative Service Officers, Deputy-Directors and Secretaries were directly

involved in identifying the CSFs of Aadhaar. Identified factors were filtered and validated by

the experts, and the final list of CSFs was selected based on the consensus of all the experts.

Details of the focus group members are shown in appendix section A.2. Since it was not

feasible to arrange a face-to-face meeting with all the group members (many of who were

very senior government officials) at a particular location because of their geographical

dispersal, a tele-conferencing focused group discussion was conducted. Finally, documents

were collated by the authors after consensus was achieved, and results were analyzed using

the BWM Method.

5.1 Data Analysis

MCDM Methods

Since this study is based on the evaluation of multiple criteria's for making a final decision,

we should use some MCDM method. There have been various MCDM methods applied in

the literature (Triantaphyllou, 2013) (Wang, Chen, & Richards, 2018) (Kubler, Robert,

Neumaier, Umbrich, & Le, 2018). MCDM allows us to evaluate numerous criteria with

varying weights. From the literature, we found multiple MCDM methods being proposed and

their applicability in various sectors. Each MCDM method has its own advantages and

disadvantages. In Table A.3 we tried to summarize some of the commonly used MCDM

methods which could also have been used in our study. We mainly focused on preferred

application area, advantages and disadvantages of a particular MCDM method.

Prioritization Approach - Best-Worst Method (BWM)

Many MCDM methods are available among those we choose recently developed MCDM

know as Best-Worst Method (Rezaei, 2015, 2016). The reason to choose BWM is because it

produces more consistent results, requires fewer data points, does not rely on complete

pairwise comparison matrix like AHP and has sophisticated pairwise comparison procedure

(Rezaei, 2015). It is also believed by the decision makers that, BWM is very close to how

they actually process and make decisions in real life scenario. Because of its simplicity and

capability to produce consistent results, this method has been used in many problems like

identification of factors that influence standard dominance in business to government data

exchange (Kaa, Janssen, & Rezaei, 2018), supplier classification (Torabi, Giahi, &

Sahebjamnia, 2016), risk assessment (Torabi et al., 2016) innovation management (Gupta &

Barua, 2016) supply chain management (Badri Ahmadi, Kusi-Sarpong, & Rezaei, 2017; Wan
Ahmad, Rezaei, Sadaghiani, & Tavasszy, 2017) , logistics performance measurement

(Rezaei, Roekel, & Tavasszy, 2018), measurement of research and development performance

(Salimi & Rezaei, 2018), and scientific output evaluation (Gupta & Barua, 2016).

Based on the academic literature and official government reports published by UIDAI from

time to time, (Ronald et al., 2017; UIDAI, 2010; Zelazny, 2012) a list of nine over-arching

goals were identified. Inputs provided by the focus group were evaluated against these goals.

The complete list of identified goals is shown is Table 4 and explained in detail in appendix

section A.2.

Table 4: Identification of Goals

Goal Definition/description Supporting References

Uniqueness Signifies that an identifier is unique in the database. (Laurent & Bouzefrane,
De-duplication achieved by using multiple attributes. 2015; UIDAI, 2010)
Enables one-to-one association between identifier
and individual.
Inclusion Every resident of the country has to have one (Jacobsen, 2012; Sharma,
Aadhaar number. 2016; UIDAI, 2010)
Convenience Getting Aadhaar should be easy in terms of (UIDAI, 2010)
enrolment, collection and modification of details for
beneficiary.
Cost- To reduce the overall costs of Aadhaar project such (Ronald et al., 2017;
optimization that it is economically feasible for both government Sharma, 2016)
and residents.
Speed To ensure coverage of target beneficiaries in a (UIDAI, 2010)
limited time frame.
Scalability To utilize such technologies and algorithms which (UIDAI, 2010; Zelazny,
are scalable and can keep up with the continuous new 2012)
enrolments in Aadhaar.
Platform To develop Aadhaar as a technology platform that (Dass, 2011; Ronald et al.,
Development can provide yes/no answer such that it could be used 2017)
in number of different domains.
Technology To adopt to information and communication (Zelazny, 2012)
Adaptation technologies which have lesser chances of becoming
obsolete in near future.
Security and The data of 1.2 billion Indian residents is stored and (Agrawal, Banerjee, &
Privacy hence the criticality of information security and Sharma, 2017; UIDAI,
privacy 2010)
7. Results

In this study, multiple criteria which are vital for each goal were identified and BWM is

applied to compute weights of these criteria. Prioritization is done at two stages, first

corresponding to each goal its design and execution choices are prioritized which depicts the

importance of each criterion and second nine identified goals of Aadhaar are prioritized to

develop a final hierarchy of goals based on their significance making it easier to follow and

replicate. It will help in taking up the most important goal and then, based on their priority

order of design and execution choices could be followed to ensure success in implementation.

Next, as an example, we have shown implementation of BWM on design choices of

uniqueness goal in appendix section A.3. Similarly prioritization of each criteria for each goal

is computed which depicts the relative importance of a particular criteria in achieving a

particular goal. Detailed results for remaining goals are added in the appendix section A.2.

In addition to prioritizing criteria for each identified goal, we have attempted to

prioritize the over-arching goals of Aadhaar. Here also BW method is used for prioritizing

these goals. From the Table 5 below it could be observed that out of nine goals Uniqueness

has the highest priority followed by Data Security and Privacy, Resident Convenience, Cost-

optimization, Speed, Inclusion, Building it as a platform, Future-proofing, and Scalability.

This prioritization could help in resource/budget allocation for each goal based on its

importance for designing similar biometric identification programs.

Table 5: Prioritization Summary of Goals of Aadhaar

Goal Weight Priority #Design CR of #Execution CR of
Criteria Design Criteria Execution
Criteria Criteria
Uniqueness of 0.294 1 3 0.090 5 0.001
IDS
Inclusion 0.059 6 3 0.134 4 0.150
Resident 0.119 3 8 0.075 6 0.070
Convenience
Cost 0.089 4 9 0.046 10 0.06
Optimisation
 Speed 0.071 5 3 0.135 9 0.064
Scalability 0.025 9 1 _ 4 0.091
Building it as 0.051 7 1 _ 6 0.084
a platform
Future- 0.044 8 1 _ 5 0.071
proofing of
technology
Data Security 0.179 2 8 0.05 7 0.079
and privacy
Where, CR is consistency ratio

Further to verify our results, we used TISM Methodology for analyzing the

relationship among the goals; and MICMAC analysis to identify the driving and dependent

power of each goal. Idea is to check if computed priorities of goals are also supported by

TISM and MICMAC analysis or not. To keep it less textual we have used shorter labels for

goals and assigned code to each goal for computation as shown in Table A.4 in appendix.

6.1 TISM/ISM

Total Interpretive Structural Modeling originated from Interpretive Structural

Modeling – a process which is employed to transform unclear and ambiguous mental models

into clear visible models (Sushil, 2012). TISM tries to address the limitation of ISM by

answering one basic question of “why”. TISM and ISM has been widely used, (Kumar et al.,

2018) used it for choosing a city for smart city project, for food logistics (Shankar, Gupta, &

Pathak, 2018), (Shibin, Gunasekaran, & Dubey, 2017) used for explaining sustainable supply

chain performance, (Shukla & Mattar, 2019) used it to identify barriers in application of

Bigdata analytics based sustainable auditing system . In this study, it is used to prioritize the

goals of Aadhaar.

We have implemented TISM methodology elaborated by (Sushil, 2012) and (V. Jain

& Raj, 2015). The first phase in TISM is to identify and define elements whose priorities’ are

to be identified. Second, contextual relationships among factors are identified via pair-wise

comparisons. Third, interpretive logic – knowledge base is developed and each pair-wise
comparison is interpreted based on directional relations that operate in a given context by

answering interpretive query “A is of higher priority than B. All detailed matrix computations

are not shown in this paper as the purpose of applying TISM is to compare the results with

BWM.

In the next step, paired comparisons in the interpretive logic—knowledge base (see

Table A.9) are translated into initial reachability matrix. If the entry in knowledge base is “Y”

then corresponding cell in the reachability matrix is marked 1 or else 0. Once initial

reachability matrix is developed it is checked for transitivity property and finally converted

into final reachability matrix (Dubey & Ali, 2014; Sushil & Sushil, 2005). Transitivity

property means if p helps q and q helps r, then p helps r also. The final reachability matrix is

shown in Table A.5 in appendix.

Next, factors are arranged into hierarchical form based on their ranking. Reachability

set (RS) and antecedent set (AS) corresponding to each factor is computed as shown in Table

A.6. Corresponding to each factor, intersection set (IS) is computed between RS and AS.

During iteration process if RS and IS are same for any factor then that factor is kept in the top

level of hierarchy. In the subsequent iterations, factors with levels assigned are removed from

the partition matrix and whole process is repeated for remaining factors. This iteration

process is continued till all factors are assigned a level (Warfield, 1974) and (Sushil, 2012).

After portioning is completed, canonical matrix is developed as shown is Table A.7 in

appendix. In our case it took six iterations to determine the level of each element. The final

partition matrix is shown in Table 6 below.

Table 6: Final Partitioning Matrix

Element Reachability Antecedents Inters- Levels
F1 {F1,F2,F4} {F1,F3,F5,F6,F7,F8,F9} {F1} II
F2 {F2} {F1,F2,F3,F5,F6,F7,F8,F9} section
{F2} I
F3 {F1,F2,F3,F4,F5,F7,F8,F9} {F3,F6} {F3} V
F4 {F4} {F1,F3,F4,F5,F6,F7,F8,F9} {F4} I
F5 {F1,F2,F4,F5} { F3,F5,F6,F7,F8,F9} {F5} III
 F6 {F1,F2,F3,F4,F5,F6,F7,F8,F9} {F6} {F6} VI
F7 {F1,F2,F4,F5,F7} {F3,F6,F7} {F7} IV
F8 {F1,F2,F4,F5,F8} {F3,F6,F8} {F8} IV
F9 {F1,F2,F4,F5,F9} {F3,F6,F9} {F9} IV
Once final partitioning matrix is generated, it is converted into graphical TISM

hierarchy model as shown in Fig 5. The TISM model indicates the prioritization and

arrangement of factors based on their significance level through a systematic computation. It

could be seen from the hierarchy model that uniqueness is the most significant goal followed

by security and privacy (bottom level of the hierarchy means most significant whereas top

most level means least significant). Speed, cost and convenience are at the same level which

depicts they are of same importance followed by less significant one, inclusion which is rated

higher than platform. The final level of TISM model is shared by two least significant goals

i.e. future-proofing and scalability.

Fig. 5: TISM Hierarchy Model

When we compare results of TISM with BWM results (see Table 7), it is found that both

results complement each other and are very similar. This similarity of results between BWM

and TISM methodology further validates the output of this research.

Table 7: Comparison of Prioritization based on BWM and TISM

Significance in Significance in
Goal BWM TISM Comparison
Rank Weight Rank Level
Uniqueness 1 0.294 1 6 Strongly Agree
Security and Privacy 2 0.179 2 5 Strongly Agree
Cost 4 0.089 3 4 Agree
Speed 5 0.071 3 4 Agree
Convenience 3 0.119 3 4 Strongly Agree
Inclusion 6 0.059 4 3 Strongly Agree
Platform 7 0.051 5 2 Strongly Agree
Future-proofing 8 0.044 6 1 Agree
Scalability 9 0.025 6 1 Strongly Agree

6.2 MICMAC

MICMAC was evolved by Duperrin and Godet in 1973 (Hu, H.-Y, Chui, S.-I, Yen, 2009).

Based on the output of ISM methodology, it categorizes list of factors into four quadrants i.e.

drivers, linkages, dependents and autonomous. Each quadrant classifies factors based on their

position, which is identified by the driving and dependence power of a particular factor.

MICMAC and its variants have been used to solve different problems like identifying barriers

of mobile-commerce adoption in small and medium enterprises (Rana, Barnard, Baabdullah,

& Rees, 2019), identification of CSF for reusable plastic packing (Gardas, Raut, & Narkhede,

2019), categorization of critical infrastructure sectors in India (Narain, Gupta, & Ojha, 2014),

analysis of obstructions in the reduction of agri-food supply chain in India (Gokarn &

Kuthambalayan, 2017), identification of reasons behind changing project management offices

(Bredillet, Tywoniak, & Tootoonchy, 2018) etc.

We used MICMAC methodology to analyze the hierarchical relationship among

factors based on their driving and dependence power. Nine factors were categorized into
three categories. It was observed that F6 (Uniqueness) and F3 (Security and Privacy) possess

higher driving power and F1 (Platform), F2 (Future-proofing), and F5 (Inclusion) possess

higher dependent power whereas F7 (Cost), F8 (Speed) and F9 (Convenience) form linkages

with somewhat balanced driving and dependent power.

Fig. 6: MICMAC Analysis of Aadhaar Goals

Where, F1 is Platform; F2 is Future-proofing; F3 is Security and Privacy; F4 is Scalability;

F5 is Inclusion; F6 is Uniqueness; F7 is Cost; F8 is Speed and F9 is Convenience
Cluster classification scatter diagram is developed based on the output of MICMAC analysis.

It differentiates between the set of goals based on their significance. From the Fig 6 above,

group of clusters is formed based on the driving and dependence strength of a particular goal.

Highest ranked goals in BW and TISM form driving factor category, and lowest ranked

goals form dependent factor category in MICMAC analysis whereas remaining goals form

linkage category. This way, MICMAC analysis also verifies the consistency, correctness and

significance of goals as computed in BWM and TISM methodologies.

8. Discussion

The success of any Biometric Identification Systems is related to the biometrics employed in

that system and how well they are performing. Purpose of the biometric system varies based

on its application, e.g., Security, Education, Sports, Healthcare, Human identification,

Government, Law enforcement, Banking, Manufacturing (Ilie-Zudor, Kemény, van

Blommestein, Monostori, & van der Meulen, 2011). In the literature section, the study on

biometrics has been done based on (i) Functional utility (ii) Technical architecture and (iii) its

process flow. In the recent past, many countries have thought of having a unique

identification identity for its citizens for providing benefits to its citizens directly, some

countries have already implemented, and some are starting now. Considering the complexity

involved in terms of budget allocation, manpower required, technological requirement, and

policy development in implementing a large-scale biometric identification program for

citizen identification at country level, a stringent action plan is needed to avoid any

unnecessary risks that might occur in the tenure of the project which could result in

substantial losses. To the best of our knowledge, we did not find any study in the literature

related to the prioritization of design and execution choices made during the development of

any biometric identification system.

Fig. 7: Prioritization and Classification of Goals

Fig 7 illustrates the final prioritization and classification hierarchy of digital identity goals.

High priority goals are labeled as drivers and less priority goals are labeled as dependents

whereas goals under linkage label have medium priorities. This prioritization could be very

helpful when implementing a biometric identification system from scratch as it depicts what

goals should be taken up first and what goals could be taken up in the later stages of the

project development.
8.1 Implications for theory

This study contributes to the theory on digital identity in a sense that it uses combination of

two theories to explain the phenomena of designing and developing a digital identity system.

This is for the first time that CSF and DT has been used together to explain the design and

development process of any digital identity system. The adoption of guidelines for

developing a digital identity system is always a good practice. Considering the

implementation and developmental cost of a typical digital ID scheme which is around £100-

250 million (Identity, 2018), in case of Aadhaar the total cost of Aadhaar project is INR

60,000 – 70,000 crores (McKinsey, 2010; Venkatanarayanan, 2018), it is essential to evaluate

each decision beforehand with utmost importance such that no significant setbacks occur

afterwards. Having a sophisticated digital ID scheme in place could save £5-10 billion by

reducing identity-related fraud and improving operational efficiency (Identity, 2018). E-

government schemes are complex and involve different actors, ambitions and perspectives

(Larsson & Grönlund, 2014). E-government schemes have multiple stakeholders, and it is

vital to give due diligence to their needs and aspirations right from the initial stages of the

project development such that the probability of project failure is minimized. Studies to

evaluate the significance of decisions taken during the design, development and

implementation phases with focus on stakeholder aspirations is said to have high impact in

the field of information and communication technology development. This study contributes

to understanding the overall developmental process of e-government projects. It adds to the

Social Construction of Technology (SCOT) study on Aadhaar (Sivamalai, 2013), that tries to

explain how perception of different stakeholders vary about Aadhaar based on the SCOT

framework. The result of this study not only supports the concept of stakeholder involvement

during e-governance project development but also suggests that the development and

implementation of project should be guided by the priorities of CSFs of that particular

project. The results of this study are also consistent with the previous research that

highlighted that confidentiality must be enforced over enrolment (this study found security

and privacy as second most important goal) (Belanche-gracia, Casaló-ariño, & Pérez-rueda,

2015; Mali & Avila-Maravilla, 2018).

Failure of UK's National ID could be considered as a validity check for this research.

One of the primary reasons for scarping this project was its cost (Travis, 2010). UK

government spent around £4.5 billion from July 2002 to February 2010 on NID and had spent

£250 million on developing it. The price of a single ID card was £30 which was considered

too expensive. The significance of cost has been detrimental for National ID and this

significance is reflected in our research as well as cost is the fourth most important goal for

an identity system. Our study strengthens the understanding surrounding the factors which

impact the outcome of such digital identity projects, even in developed economies.

Considering the digital identity as domain, no such study was found from the literature which

has prioritized the goals of Aadhaar and decisions taken during design and development of

Aadhaar system. With a focus on identifying linkages among goals of Aadhaar and multiple

decisions evaluated during the design and implementation of the Aadhaar system, this study

is unique in its own way.

The study is unique for its methodological contributions as well. From the research

methodology point of view, the main contribution of this study is the combination and

application of three different methodologies i.e. BWM, TISM and MICMAC, on a single

large scale project and to analyze the process and factors taken into consideration for decision

making in such a massive and critical national project. Also, the knowledge obtained through

the method and technique implemented for collecting data may be helpful in other studies on

analyzing large scale government initiatives. The integrated usage of three different
methodologies also ensures higher rigor in the methodology which brings more confidence

on the reliability and validity of the outcome.

Overall findings from three different methods reveal that in an identity scheme

uniqueness of an entity and privacy and security of individual’s data is of utmost importance

and is the main driving factors of the whole scheme. The inclusion of entire population,

building identity project as a platform rather than a single standalone system, making project

capable of withholding dynamic technological innovations and a mechanism to scale project

as and when needed are those objectives which are vital in an identity system but could be

taken up in later stages of the project development. These four objectives are highly

dependent on drivers and linkages, and hence it is logical to focus on drivers and linkages

first as shown in prioritization hierarchy diagram. Linkages which comprise of cost, speed

and convenience act as intermediaries between drivers and dependent factors and have higher

priority than dependents. Based on the results we recommend to follow priority order

however in case of linkages, three objectives, i.e. cost, speed and convenience could be

shuffled if situation demands for it.

8.2 Implications for practice

Aadhaar has received much attention across the globe ever since UN recommended to

provide the legal identity for all in its SDG 16 (UNGA, 2016). Aadhaar being a massive

biometric project at present with more than 1.2 billion Aadhaar numbers issued, it has

become a system which other countries may contemplate replication while implementing a

biometric digital identity system. To start any such critical mega-project requires a lot of

planning while focusing on budget estimation, implementation policy, and development plan

and to identify various types of risks associated with the project at each stage. For developing

nations, it may not be feasible to reinvent the wheel and conduct this analysis because of

time, budget and expertise constraints. This research is intended to contribute in this space by
bringing out both design and implementation factors for public policy makers of future digital

identity systems. It will act as a reference for all those nations who are working on UN’s

SDG 16 for providing legal identity to all its individuals. It will be helpful in taking up a

biometric identification program in a systematic manner by giving specific attention to high

priority tasks. It will help concerned nations to save significant amount of time and money

which otherwise would have been mandatory for conducting a pre-launch analysis. It will

also enable governments to identify different types of risks associated at various stages of

program and will allow them to have mitigation measures thereby increasing chances of

having a successful, efficient project in place. In this study, we have shown how important

each criterion is for accomplishing a particular goal. We also found how priority of same

criterion varies among different goals. Further, we also illustrated through a case study on

Aadhaar how BWM could be used in real-world complex decision-making problems.

9. Conclusion

We conclude our study of prioritizing goals of Aadhaar by verifying our results with the help

of two more methodologies- TISM and MICMAC analysis. Original prioritization results are

also supported by both TISM and MICMAC results which further strengthen our findings.

Digital identities, in general, have the capacity for both bliss and misery. A well planned

digital identity system, having necessary measures in place that can address issues like

security, privacy, inclusion and citizen empowerment could unfold remarkable economic

values. In this work, we conducted a detailed study to identify the primary goals that are must

for any biometric identification system. India's biometric identification program- Aadhaar has

been used as a case study and has been analyzed from the perspective of DT in this study.

After prioritizing design goals of Aadhaar using Best-Worst Method and verified using TISM

and MICMAC, it was observed that three clusters of goals were formed. All three clusters are
critical for any biometric identification program but with varying priorities. First cluster (i.e.

drivers) has the highest priority and must be dealt on priority in the initial stages of the

program, second cluster (i.e. linkages) is the second most significant one and could be taken

right after the first cluster, and third cluster (i.e. dependents) which has comparatively less

priority as compared to drivers and linkages and could be taken up in the later stages of the

program development. This study also tests the application of BWM method on a real-world

complex problem for evaluating the opinion of focus group members.

Since this research is limited to only one case study –Aadhaar, we do not claim priorities

of goals as an absolute one. We hope to inspire researchers across the world to conduct

similar studies on other national biometric identity projects and develop a universal

prioritization hierarchy. Future research direction is to verify the prioritized goals on some

more existing biometric identification system and use the results of prioritization in designing

the real-world biometric identification system. Also, security and privacy have emerged as

the second most crucial goal in this study. However, frequent complaints related to the

security and privacy of Aadhaar highlights that something is still lacking. This gap could be

another exciting area to explore.

Funding

This research did not receive any specific grant from funding agencies in the public,

commercial, or not-for-profit sectors.

References:
Abate, A. F., Marcialis, G. L., Poh, N., & Sansone, C. (2019). Introduction to the special

issue on robustness, security and regulation aspects in current biometric systems

(RSRA-BS). Pattern Recognition Letters, 126, 1–2.

https://doi.org/10.1016/j.patrec.2019.07.004

Agogué, M., & Kazakçi, A. (2014). 10 Years of C–K Theory: A Survey on the Academic and

Industrial Impacts of a Design Theory. In An Anthology of Theories and Models of

 Design (pp. 219–235). London: Springer London. https://doi.org/10.1007/978-1-4471-

6338-1_11

Agrawal, S., Banerjee, S., & Sharma, S. (2017). Privacy and Security of Aadhaar : A

Computer Science Perspective, 52(37), 1–23.

Akhtar Shareef, M., Kumar, V., Kumar, U., & Dwivedi, Y. (2014). Factors affecting citizen

adoption of transactional electronic government. Journal of Enterprise Information

Management, 27(4), 385–401. https://doi.org/10.1108/JEIM-12-2012-0084

Aladwani, A. M. (2016). Corruption as a source of e-Government projects failure in

developing countries: A theoretical exposition. International Journal of Information

Management, 36(1), 105–112. https://doi.org/10.1016/j.ijinfomgt.2015.10.005

Anil, K. J., & Sharathchandra, P. (2001). A Touch of Money. IEEE Spectrum, 4123(March),

1–9.

Anthopoulos, L., Reddick, C. G., Giannakidou, I., & Mavridis, N. (2016). Why e-government

projects fail? An analysis of the Healthcare.gov website. Government Information

Quarterly, 33(1), 161–173. https://doi.org/10.1016/j.giq.2015.07.003

Arutyunov, V. V., & Natkin, N. S. (2010). Comparative analysis of biometric systems for

information protection. Scientific and Technical Information Processing, 37(2), 87–93.

https://doi.org/10.3103/S0147688210020012

Badri Ahmadi, H., Kusi-Sarpong, S., & Rezaei, J. (2017). Assessing the social sustainability

of supply chains using Best Worst Method. Resources, Conservation and Recycling,

126, 99–106. https://doi.org/https://doi.org/10.1016/j.resconrec.2017.07.020

Bai, C., & Sarkis, J. (2013). A grey-based DEMATEL model for evaluating business process

management critical success factors. International Journal of Production Economics,

146(1), 281–292. https://doi.org/10.1016/j.ijpe.2013.07.011

Baichoo, S., Heenaye-Mamode Khan, M., Bissessur, P., Pavaday, N., Boodoo-Jahangeer, N.,
 & Purmah, N. R. (2018). Legal and ethical considerations of biometric identity card:

Case for Mauritius. Computer Law & Security Review, 34(6), 1333–1341.

https://doi.org/10.1016/j.clsr.2018.08.010

Bálint, T., & Bucko, J. (2013). Comparative Analysis of Handwritten, Biometric and Digital

Signature. International Review of Social Sciences and Humanities, 4(2).

Bank, W. (2016). Population of India. Retrieved from

https://data.worldbank.org/indicator/SP.POP.TOTL

BankWorld. (2016). World Development Report 2016: Digital Dividends. The World Bank.

https://doi.org/10.1596/978-1-4648-0671-1

Barnwal, P. (2015). Curbing Leakage in Public Programs with Direct Benefit Transfers

Evidence from India ’ s Fuel Subsidies and Black Markets. Working Paper, (November

2014).

BBC News. (2002). Doubt cast on fingerprint security. Retrieved December 20, 2017, from

http://news.bbc.co.uk/2/hi/science/nature/1991517.stm

Belanche-gracia, D., Casaló-ariño, L. V, & Pérez-rueda, A. (2015). Determinants of multi-

service smartcard success for smart cities development : A study based on citizens ’

privacy and security perceptions. Government Information Quarterly, 32(2), 154–163.

https://doi.org/10.1016/j.giq.2014.12.004

Ben Ayed, G. (2014). Architecting User-Centric Privacy-as-a-Set-of-Services. Cham:

Springer International Publishing. https://doi.org/10.1007/978-3-319-08231-8

Bergeron, F., & Bégin, C. (1989). The Use of Critical Success Factors in Evaluation of

Information Systems: A Case Study. Journal of Management Information Systems, 5(4),

111–124. https://doi.org/10.1080/07421222.1989.11517842

Bouchrika, I. (2017). A Survey of Using Biometrics for Smart Visual Surveillance: Gait

Recognition. In Surveillance in Action (pp. 3–23). Springer. Retrieved from

 https://link.springer.com/chapter/10.1007/978-3-319-68533-5_1

Boynton, A., & Zmud, R. (1984). An Assessment of Critical Success Factors. Sloan

Management Review, 25(4), 17.

Bredillet, C., Tywoniak, S., & Tootoonchy, M. (2018). ScienceDirect Why and how do

project management of fi ces change ? A structural analysis approach. International

Journal of Project Management, 36(5), 744–761.

https://doi.org/10.1016/j.ijproman.2018.04.001

Buchanan, R. (2019). Systems Thinking and Design Thinking: The Search for Principles in

the World We Are Making. She Ji: The Journal of Design, Economics, and Innovation,

5(2), 85–104. https://doi.org/10.1016/j.sheji.2019.04.001

Burgess, S. (2010). The Use of Focus Groups in Information Systems Research. The

International Journal of Interdisciplinary Social Sciences: Annual Review, 5(2), 57–68.

https://doi.org/10.18848/1833-1882/CGP/v05i02/51567

Caroline Tynan, A., & Drayton, J. L. (1988). CONDUCTING FOCUS GROUPS — A

GUIDE FOR FIRST‐TIME USERS. Marketing Intelligence & Planning, 6(1), 5–9.

https://doi.org/10.1108/eb045757

Chandra, A., & Calderon, T. (2005). Challenges and Constraints to the Diffusion of

Biometrics in Information Systems. In Communications of the ACM (p. 48).

Coule, T. (2013). Theories of knowledge and focus groups in organization and management

research. Qualitative Research in Organizations and Management: An International

Journal, 8(2), 148–162. https://doi.org/10.1108/QROM-09-2011-1006

Dass, R. (2011). Unique Identification for Indians: A Divine Dream or a Miscalculated

Heroism?, 36(1), 1–14. https://doi.org/10.1177/0256090920110101

Daugman, J. G. (1993). High confidence visual recognition of persons by a test of statistical

independence. IEEE Transactions on Pattern Analysis and Machine Intelligence,

 15(11), 1148–1161. https://doi.org/10.1109/34.244676

Dixon, P. (2017). A Failure to Do No Harm – India ’ s Aadhaar biometric ID program and its

inability to protect privacy in relation to measures in Europe and the U . S . Health and

Technology. https://doi.org/10.1007/s12553-017-0202-6

Dubey, R., & Ali, S. S. (2014). Identification of flexible manufacturing system dimensions

and their interrelationship using total interpretive structural modelling and fuzzy

MICMAC analysis. Global Journal of Flexible Systems Management, 15(2), 131–143.

https://doi.org/10.1007/s40171-014-0058-9

Elkadi, H. (2013). Success and failure factors for e-government projects: A case from Egypt.

Egyptian Informatics Journal, 14(2), 165–173. https://doi.org/10.1016/j.eij.2013.06.002

Epstein, C. (2008). Embodying Risk:Using Biometrics To Protect the Borders. In Risk and

the War on Terror (pp. 178–193). Routledge. Retrieved from

https://www.academia.edu/2420067/_Embodying_Risk_Using_Biometrics_to_Protect_t

he_Borders_in_L_Amoore_and_M_de_Goede_ed._2008_Risk_and_the_War_on_Terro

r_Routledge_London

Esfahlan, E. H., & Valilai, O. F. (2019). A Knowledge Oriented Framework to Enable New

Service Development Using Axiomatic Design Theory: Focusing on Hotel Business

Models. In 2019 15th Iran International Industrial Engineering Conference (IIIEC) (pp.

267–273). IEEE. https://doi.org/10.1109/IIIEC.2019.8720620

Fenu, G., Marras, M., & Boratto, L. (2018). A multi-biometric system for continuous student

authentication in e-learning platforms. Pattern Recognition Letters, 113, 83–92.

https://doi.org/10.1016/j.patrec.2017.03.027

Gal, U., & Whitley, E. A. (2011). Special Issue on Information Systems, Identity and

Identification. European Journal of Information Systems (EJIS), 1(1), 1–2.

Gardas, B. B., Raut, R. D., & Narkhede, B. (2019). Identifying critical success factors to
 facilitate reusable plastic packaging towards sustainable supply chain management.

Journal of Environmental Management, 236(January), 81–92.

https://doi.org/10.1016/j.jenvman.2019.01.113

Gelb, A., & Clark, J. (2013). Performance Lessons from India ’ s Universal Identification

Program CGD Policy Paper 020 May 2013, (May).

GOI. (2016). Elderly in India. Retrieved from

http://mospi.nic.in/sites/default/files/publication_reports/ElderlyinIndia_2016.pdf

Gokarn, S., & Kuthambalayan, T. S. (2017). Analysis of challenges inhibiting the reduction

of waste in food supply chain. Journal of Cleaner Production, 168, 595–604.

https://doi.org/10.1016/j.jclepro.2017.09.028

Goldkuhl, G. (2004). Design Theories in Information Systems – A Need for Multi-

Grounding. Journal of Information Technology Theory and Application (JITTA), 6(2),

59–72. Retrieved from

https://aisel.aisnet.org/cgi/viewcontent.cgi?article=1127&context=jitta

Gregor, S., & Jones, D. (2007). The Anatomy of a Design Theory. Journal of the Association

for Information Systems, 8(5), 312–335. Retrieved from

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.232.743&rep=rep1&type=pdf

Gupta, H., & Barua, M. K. (2016). Identifying enablers of technological innovation for Indian

MSMEs using best–worst multi criteria decision making method. Technological

Forecasting and Social Change, 107, 69–79.

https://doi.org/https://doi.org/10.1016/j.techfore.2016.03.028

HALL, R. R. (2001). Prototyping for usability of new technology. International Journal of

Human-Computer Studies, 55(4), 485–501. https://doi.org/10.1006/ijhc.2001.0478

Hartman, J. (2004). Using Focus Groups to Conduct Business Communication Research.

Journal of Business Communication, 41(4), 402–410.

 https://doi.org/10.1177/0021943604267775

Hedström, K., Karlsson, F., & Söderström, F. (2016). Challenges of introducing a

professional eID card within health care. Transforming Government: People, Process

and Policy, 10(1), 26–46. https://doi.org/10.1108/TG-10-2013-0044

Hoang, B., & Caudill, A. (2012). Biometrics. IEEE Emerging Technology Portal, 1–3.

Hodeghatta, U., & Nayak, R. (2014). Physical Security and Biometrics. In The InfoSec

Handbook (pp. 293–306). SpringerLink. https://doi.org/https://doi.org/10.1007/978-1-

4302-6383-8_14

Howlett, M. (2014). From the ‘old’ to the ‘new’ policy design: design thinking beyond

markets and collaborative governance. Policy Sciences, 47(3), 187–207.

https://doi.org/10.1007/s11077-014-9199-0

Hu, H.-Y, Chui, S.-I, Yen, T.-M. (2009). Modified IPA for order-winner criteria

improvement: A MICMAC approach. Journal of Applied Sciences, 9(21), 3792–3803.

Identity, I. (2018). Digital Identity in the UK : The cost of doing nothing, (April).

Ilie-Zudor, E., Kemény, Z., van Blommestein, F., Monostori, L., & van der Meulen, A.

(2011). A survey of applications and requirements of unique identification systems and

RFID techniques. Computers in Industry, 62(3), 227–252.

https://doi.org/10.1016/j.compind.2010.10.004

Irani, Z., Kamal, M., Axelsson, U. M., Karin, Wihlborg, E., & Janssen, M. (2016). Editorial.

Transforming Government: People, Process and Policy, 10(1), 2–7.

https://doi.org/10.1108/TG-01-2016-0009

ISO. (2011). Information technology -- Security techniques -- A framework for identity

management -- Part 1: Terminology and concepts. Retrieved from

https://www.iso.org/standard/57914.html

ITU-T. (2010). X.1252 - Baseline identity management terms and definitions. Itu-T X-Series
 Recommendations Data Networks, Open System Communications and Security, ITU-T

X.1252 (04/2010).

Jacobsen, E. K. U. (2012). Unique Identification : Inclusion and surveillance in the Indian

biometric assemblage. https://doi.org/10.1177/0967010612458336

Jain, A K, Ross, A., & Prabhakar, S. (2004). An introduction to biometric recognition. IEEE

Transactions on Circuits and Systems for Video Technology, 14(1), 4–20.

https://doi.org/10.1109/TCSVT.2003.818349

Jain, Anil K., Nandakumar, K., & Ross, A. (2016). 50 years of biometric research:

Accomplishments, challenges, and opportunities. Pattern Recognition Letters, 79, 80–

105. https://doi.org/10.1016/j.patrec.2015.12.013

Jain, V., & Raj, T. (2015). Modeling and analysis of FMS flexibility factors by TISM and

fuzzy MICMAC. International Journal of Systems Assurance Engineering and

Management, 6(3), 350–371. https://doi.org/10.1007/s13198-015-0368-0

Kaa, G. Van De, Janssen, M., & Rezaei, J. (2018). Technological Forecasting & Social

Change Standards battles for business-to-government data exchange : Identifying

success factors for standard dominance using the Best Worst Method. Technological

Forecasting & Social Change, 137(July), 182–189.

https://doi.org/10.1016/j.techfore.2018.07.041

Krueger, R., & Casey, M. (1994). Focus Groups: A Practical Guide for Applied Research.

Sage Publications.

Kubler, S., Robert, J., Neumaier, S., Umbrich, J., & Le, Y. (2018). Comparison of metadata

quality in open data portals using the Analytic Hierarchy Process. Government

Information Quarterly, 35(1), 13–29. https://doi.org/10.1016/j.giq.2017.11.003

Kumar, H., Singh, M. K., & Gupta, M. P. (2018). A policy framework for city eligibility

analysis: TISM and fuzzy MICMAC-weighted approach to select a city for smart city
 transformation in India. Land Use Policy, 82(February 2018), 375–390.

https://doi.org/10.1016/j.landusepol.2018.12.025

Lambert, D. M., & Enz, M. G. (2017). Issues in Supply Chain Management: Progress and

potential. Industrial Marketing Management, 62, 1–16.

https://doi.org/10.1016/j.indmarman.2016.12.002

Larsson, H., & Grönlund, Å. (2014). Future-oriented eGovernance : The sustainability

concept in eGov research , and ways forward. Government Information Quarterly,

31(1), 137–149. https://doi.org/10.1016/j.giq.2013.07.004

Laurent, M., & Bouzefrane, S. (2015). Digital Identity Management. ISTE Press and

Elsevier. https://doi.org/10.1007/978-3-319-08231-8

Le Masson, P., Hatchuel, A., Le Glatin, M., & Weil, B. (2019). Designing Decisions in the

Unknown: A Generative Model. European Management Review, 16(2), 471–490.

https://doi.org/10.1111/emre.12289

Lentner, G. M., & Parycek, P. (2016). Electronic identity (eID) and electronic signature

(eSig) for eGovernment services – a comparative legal study. Transforming

Government: People, Process and Policy, 10(1), 8–25. https://doi.org/10.1108/TG-11-

2013-0047

Li, X., Sun, S. X., Chen, K., Fung, T., & Wang, H. (2015). Design Theory for Market

Surveillance Systems. Journal of Management Information Systems, 32(2), 278–313.

https://doi.org/10.1080/07421222.2015.1063312

Mali, N. V., & Avila-Maravilla, M. A. (2018). Convergence or Conflict? Proceedings of the

11th International Conference on Theory and Practice of Electronic Governance -

ICEGOV ’18, 443–448. https://doi.org/10.1145/3209415.3209487

Markus, M. L., Majchrzak, A., & Gasser, L. (2002). A Design Theory for Systems That

Support Emergent Knowledge Processes. MIS Quarterly, 26(3), 179–212. Retrieved

 from https://www.jstor.org/stable/pdf/4132330.pdf

Marohn, D. (2006). Biometrics in healthcare. Biometric Technology Today, 14(9), 9–11.

https://doi.org/https://doi.org/10.1016/S0969-4765(06)70592-6

Mauceri A.J. (1965). Feasibility Study Of Personnel Identification By Signature Verification.

Retrieved from http://www.dtic.mil/docs/citations/AD0617615

May, P.J. (2003). Policy design and implementation. In Handbook of public administration

(p. 233).

McKinsey. (2010). Inclusive growth and financial security. Retrieved from

http://ccmrm.org/wp-content/uploads/2015/05/McKinsey-2010-inclusive-growth-

report.pdf

McKinsey. (2019). Digital Identification : A Key To Inclusive Growth. Retrieved from

https://www.mckinsey.com/~/media/mckinsey/featured insights/innovation/the value of

digital id for the global economy and society/mgi-digital-identification-a-key-to-

inclusive-growth.ashx

Mcwaters, R. J. (2016). A Blueprint for Digital Identity The Role of Financial Institutions in

Building Digital Identity. World Economic Forum, Future of Financial Services Series,

(August), 1–108. Retrieved from

http://www3.weforum.org/docs/WEF_A_Blueprint_for_Digital_Identity.pdf

Meier, A., & Terán, L. (2019). Special Issue on eGovernment Research, Management, and

Innovation.

Melin, U., Axelsson, K., & Söderström, F. (2016). Managing the development of e-ID in a

public e-service context. Transforming Government: People, Process and Policy, 10(1),

72–98. https://doi.org/10.1108/TG-11-2013-0046

Mintrom, M., & Luetjens, J. (2016). Design Thinking in Policymaking Processes:

Opportunities and Challenges. Australian Journal of Public Administration, 75(3), 391–

 402. https://doi.org/10.1111/1467-8500.12211

Morgan, D. (1996). Focus Groups. Annual Review of Sociology, 22, 129–152.

Narain, A., Gupta, M. P., & Ojha, A. (2014). Identifying critical infrastructure sectors and

their dependencies : An Indian scenario. International Journal of Critical Infrastructure

Protection, 7(2), 71–85. https://doi.org/10.1016/j.ijcip.2014.04.003

NIST. (2013). Standards for Biometric Technologies. Retrieved July 21, 2019, from

https://www.nist.gov/speech-testimony/standards-biometric-technologies

Nunno, R. M. (2000). Electronic Signatures : Technology Developments and Legislative

Issues. RterlyGovernment Information Qua, 17(4), 395–401.

O.Nyumba, T., Wilson, K., Derrick, C. J., & Mukherjee, N. (2018). The use of focus group

discussion methodology: Insights from two decades of application in conservation.

Methods in Ecology and Evolution, 9(1), 20–32. https://doi.org/10.1111/2041-

210X.12860

OECD. (2018). Case Study-Aadhaar. Retrieved from https://www.oecd.org/gov/innovative-

government/India-case-study-UAE-report-2018.pdf

Olson, E. (2015). Personal Identity. In Stanford Encyclopedia of Philosophy. Retrieved from

https://plato.stanford.edu/entries/identity-personal/

P.Down, M., & J.Sands, R. (2004). Biometrics: An Overview of the Technology,Challenges

and Control Considerations. Information Systems Control Journal, 4(11–4).

Pagnin, E., & Mitrokotsa, A. (2017). Privacy-Preserving Biometric Authentication:

Challenges and Directions. Security and Communication Networks, 2017, 1–9.

https://doi.org/10.1155/2017/7129505

Phillips, P. J., Flynn, P. J., & Bowyer, K. W. (2017). Lessons from collecting a million

biometric samples. Image and Vision Computing, 58(Supplement C), 96–107.

https://doi.org/https://doi.org/10.1016/j.imavis.2016.08.004
Pluchinotta, I., Kazakçi, A. O., Giordano, R., & Tsoukiàs, A. (2019). Design Theory for

Generating Alternatives in Public Decision Making Processes. Group Decision and

Negotiation, 28(2), 341–375. https://doi.org/10.1007/s10726-018-09610-5

Pruzansky, S. (1963). Pattern‐Matching Procedure for Automatic Talker Recognition. The

Journal of the Acoustical Society of America, 35(3), 354–358.

https://doi.org/10.1121/1.1918467

Rana, N. P., Barnard, D. J., Baabdullah, A. M. A., & Rees, D. (2019). International Journal of

Information Management Exploring barriers of m-commerce adoption in SMEs in the

UK : Developing a framework using ISM. International Journal of Information

Management, 44(June 2018), 141–153. https://doi.org/10.1016/j.ijinfomgt.2018.10.009

Rana, N. P., Dwivedi, Y. K., & Williams, M. D. (2013). Analysing challenges, barriers and

CSF of egov adoption. Transforming Government: People, Process and Policy, 7(2),

177–198. https://doi.org/10.1108/17506161311325350

Rezaei, J., Roekel, W. S. Van, & Tavasszy, L. (2018). Measuring the relative importance of

the logistics performance index indicators using Best Worst Method. Transport Policy,

68(March), 158–169. https://doi.org/10.1016/j.tranpol.2018.05.007

Ronald, A., Elizabeth, S. B., Noopur, S., & Neil, S. B. (2017). State of aadhaar report 2016-

17.

Salimi, N., & Rezaei, J. (2018). Evaluating firms’ R&D performance using best worst

method. Evaluation and Program Planning, 66(September 2017), 147–155.

https://doi.org/10.1016/j.evalprogplan.2017.10.002

Segovia, A. I., Álvaro, D., & Enríquez, M. (2018). Digital Identity : the current state of

affairs. BBVA research.

Seltsikas, P., & O’Keefe, R. M. (2010). Expectations and outcomes in electronic identity

management: the role of trust and public value. European Journal of Information
 Systems, 19(1), 93–103. https://doi.org/10.1057/ejis.2009.51

Shah, M. H., Braganza, A., & Morabito, V. (2007). A survey of critical success factors in e-

Banking: an organisational perspective. European Journal of Information Systems,

16(4), 511–524. https://doi.org/10.1057/palgrave.ejis.3000693

Shankar, R., Gupta, R., & Pathak, D. K. (2018). Modeling critical success factors of

traceability for food logistics. Transportation Research Part E, 119(August 2017), 205–

222. https://doi.org/10.1016/j.tre.2018.03.006

Sharma, R. S. (2016). UIDAI ’ s Public Policy Innovations, (176), 1–19.

Shibin, K. T., Gunasekaran, A., & Dubey, R. (2017). Explaining sustainable supply chain

performance using a total interpretive structural modeling approach. Sustainable

Production and Consumption, 12(July), 104–118.

https://doi.org/10.1016/j.spc.2017.06.003

Shukla, M., & Mattar, L. (2019). Computers & Industrial Engineering Next generation smart

sustainable auditing systems using Big Data Analytics : Understanding the interaction of

critical barriers. Computers & Industrial Engineering, 128(May 2018), 1015–1026.

https://doi.org/10.1016/j.cie.2018.04.055

Simon, H. A. (1996). The Sciences of the Artificial (Third edit). London: MIT Press.

Retrieved from

https://monoskop.org/images/9/9c/Simon_Herbert_A_The_Sciences_of_the_Artificial_3

rd_ed.pdf

Singh, S., Cabraal, A., Demosthenous, C., Astbrink, G., & Furlong, M. (2007). Password

sharing: implications for security design on social practice. In Proceedings of the

SIGCHI conference on Human factors in computing systems - CHI ’07 (p. 895). New

York, New York, USA: ACM Press. https://doi.org/10.1145/1240624.1240759

Sivamalai, L. (2013). Using the lens of “social construction of technology” to understand the
 design and implementation of aadhaar (UID) project. IFIP Advances in Information and

Communication Technology, 402, 633–638. https://doi.org/10.1007/978-3-642-38862-

0_45

Spagnoletti, Paolo, Resca, A., & Lee, G. (2015). A design theory for digital platforms

supporting online communities: a multiple case study. Journal of Information

Technology, 30(4), 364–380.

Sushil. (2012). Interpreting the interpretive structural model. Global Journal of Flexible

Systems Management, 13(2), 87–106. https://doi.org/10.1007/S40171-012-0008-3

Sushil, & Sushil. (2005). Interpretive Matrix: A Tool to Aid Interpretation of Management

and Social Research. Global Journal of Flexible Systems Management, 6(2), 27–30.

Torabi, S. A., Giahi, R., & Sahebjamnia, N. (2016). An enhanced risk assessment framework

for business continuity management systems. Safety Science, 89, 201–218.

https://doi.org/https://doi.org/10.1016/j.ssci.2016.06.015

Trauring, M. (1963). Automatic comparison of finger-ridge patterns.

https://doi.org/10.1038/197938a0

Travis, A. (2010). ID cards scheme to be scrapped within 100 days. The Guardian Weekly.

Retrieved from https://www.theguardian.com/politics/2010/may/27/theresa-may-

scrapping-id-cards

Triantaphyllou, E. (2013). Multi-criteria Decision Making Methods: A Comparative Study

(Vol. 44). Boston, MA: Springer US. https://doi.org/10.1007/978-1-4757-3157-6

UIDAI. (2009). Biometrics Design Standards For UID Applications. Retrieved from

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.189.2453&rep=rep1&type=pd

UIDAI. (2010). UIDAI Strategy Overview Creating a Unique Identity Number for Every

Resident in India, 1–45. Retrieved from

 http://eprints.cscsarchive.org/367/1/Strategy_Overveiw-001.pdf

UIDAI. (2019). Aadhaar Dashboard. Retrieved February 3, 2019, from

https://uidai.gov.in/aadhaar_dashboard/

Uludag, U., Pankanti, S., Prabhakar, S., & Jain, A. K. (2004). Biometric cryptosystems:

issues and challenges. Proceedings of the IEEE, 92(6), 948–960.

https://doi.org/10.1109/JPROC.2004.827372

UN. (2016). SDG. Retrieved February 8, 2019, from

https://www.un.org/sustainabledevelopment/peace-justice/

UNGA. (2016). SDG. Retrieved from

https://www.un.org/development/desa/disabilities/envision2030-goal16.html

Venkatanarayanan, A. (2018). Aadhaar enrolment costs. Retrieved July 4, 2019, from

https://medium.com/karana/aadhaar-enrolment-costs-bc17f0d30018

Walls, J. G., Widmeyer, G. R., & El Sawy, O. A. (1992). Building an Information System

Design Theory for Vigilant EIS. Information Systems Research, 3(1), 36–59.

https://doi.org/10.1287/isre.3.1.36

Wan Ahmad, W. N. K., Rezaei, J., Sadaghiani, S., & Tavasszy, L. A. (2017). Evaluation of

the external forces affecting the sustainability of oil and gas supply chain using Best

Worst Method. Journal of Cleaner Production, 153, 242–252.

https://doi.org/https://doi.org/10.1016/j.jclepro.2017.03.166

Wang, D., Chen, C., & Richards, D. (2018). A prioritization-based analysis of local open

government data portals : A case study of Chinese province-level governments.

Government Information Quarterly, 35(4), 644–656.

https://doi.org/10.1016/j.giq.2018.10.006

Warfield, J. N. (1974). Developing Subsystem Matrices in Structural Modeling. IEEE

Transactions on Systems, Man and Cybernetics, SMC-4(1), 74–80.

 https://doi.org/10.1109/TSMC.1974.5408523

Whitley, E. A., Gal, U., & Kjaergaard, A. (2014). Who do you think you are? A review of the

complex interplay between information systems, identification and identity. European

Journal of Information Systems, 23(1), 17–35. https://doi.org/10.1057/ejis.2013.34

Wilkinson, S. (1998). Focus group methodology: a review. International Journal of Social

Research Methodology, 1(3), 181–203.

https://doi.org/10.1080/13645579.1998.10846874

Zelazny, F. (2012). The Evolution of India ’ s UID Program Lessons Learned and

Implications for Other Developing Countries CGD Policy Paper 008, (August).