Procedia Engineering 159 ( 2016 ) 199 – 209

Available online at www.sciencedirect.com

1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).Peer-review under responsibility of the Organizing Committee of HumTech2016doi: 10.1016/j.proeng.2016.08.159

ScienceDirect

* Corresponding author. Tel.: +1-415-757-7877. E-mail address: zalzala@communitytracks.org

Humanitarian Technology: Science, Systems and Global Impact 2016, HumTech2016, 7-9 June 2016, Massachusetts, USA

Disruption of things: a model to facilitate adoption of IoT-based innovations by the urban poor

Abhimanyu Roya,c, Ali M S Zalzalab,c,*, Alok Kumara aInstitute of Management Technology, Ghaziabad, India

bInstitute of Management Technology, Dubai, UAE cCommunity Tracks, USA

Abstract

This study examines the adoption of the Internet of Things (IoT) based innovations by urban poor communities. In recent years, IoT applications in social networking, smart cities and m-health have been explored, but the social acceptance of such applications remain largely unknown. A consumer segment on which IoT can have a major positive impact is the urban poor, where IoT can provide access to services such as healthcare, education and food security. However, to facilitate adoption among the poor, IoT-based innovations must incorporate the unique characteristics of this segment viz. low levels of technology awareness, social acceptance and consumer need. The Disruption of Things refers to the process of distribution the innovations that are based on IoT and replacing existing market leaders and prevalent systems. The study was conducted in four stages – a literature review, a survey with the target users, interviews with experts (both technological and sociological) and a usability test with a prototype technology system. The results from the surveys, interviews and usability tests were used to develop a model for adoption of IoT-based innovations by the urban poor. The model identifies five sources of innovation – nutrition, healthcare, employment, education and finances. Based on these sources, a participative design process is undertaken. Once developed the innovation must provide excellent service based on three parameters – benefits (value of using the system to the users), support provided to the users, and training/instructions given to them regarding system use. Satisfied system’s users would then be leveraged through three channels – advertising, social media and word-of-mouth, in order to spur greater adoption of the innovation among the urban poor. Accomplishing these systematic processes would enable the Disruption of Things – replacement of existing market leaders by an IoT-based innovation. © 2016 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the Organizing Committee of HumTech2016

Keywords: IoT; urban poor; innovation

1. Scope and Benefits

This study aims to develop a model that facilitates the adoption of IoT-based innovations by the urban poor. IoT is a global information infrastructure that enables advanced services by interconnecting devices based on existing and evolving interoperable information and communication technologies [1]. IoT facilitates communication forms beyond the traditional human-human to human-thing and thing-thing (also referred to as M2M). Radio Frequency Identification (RFID) is believed to be a foundation for IoT [2].

While the original intent of IoT was to increase supply chain efficiencies for organizations [3], more recently potential IoT applications in social networking [4], smart cities [5] and m-health [6] have been explored. However, all such potential applications face a road block in the form of societal acceptance. Sanchez [7] noted that social acceptance was a major challenge in the implementation of smart cities. Coughlan [8] considered that social acceptance of IoT applications depended on a number of human factors such as sociology, psychology and human-computer interaction.

© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).Peer-review under responsibility of the Organizing Committee of HumTech2016

brought to you by COREView metadata, citation and similar papers at core.ac.uk

provided by Elsevier - Publisher Connector

200 Abhimanyu Roy et al. / Procedia Engineering 159 ( 2016 ) 199 – 209

One consumer segment on which IoT-based innovations can have a major impact is the urban poor. On the face of it, the urban poor does not represent a profitable segment. Consumers in this segment have relatively small incomes – most of them live on less than $2 a day [9] – and exhibit low levels of technology adoption [10]. Contrary to this line of thinking, the use of IoT for providing access to services for the poor have been considered for food security [11], education [12] and healthcare [13,14]. These examples reflect the major area where IoT can affect the lives of the urban poor – providing access to services that they otherwise would not have access.

Studies show that providing access to services such as healthcare, education and food security can have major benefits for the urban poor [15,16,17]. However, unlike other consumer segments, a model that facilitates the adoption of IoT-based innovations mandates accounting for the unique characteristics inherent to this segment. For this reason, there is a need for a model that reconciles technological innovation and user adoption to facilitate the replacement of market leaders by IoT-based innovations. This is the essence of the Disruption of Things – the process of distribution of innovations that are based on IoT and replacing existing market leaders and prevalent systems. Such a model, though specific to the urban poor would not be a unique development as frameworks that facilitate the adoption of eCommerce [18], online banking services [19] and mobile value-added services [20] have been proposed in the past.

This paper is structured as follows. Section 2 defines the research objectives, while methodology is explained in Section 3 and including instruments and data sources. The sources of innovations defined through analysis are reported in the results Section 4, and we included supporting literature survey for each source area in Section 5. Innovations adoption model is deduced from our analysis and reported in Section 6, and brief description of two technology solutions piloted accordingly are included in Section 7. Limitations and on-going work are given in Section 8, with conclusions drawn in Section 9.

2. Approach and Objectives

Heeks [21] noted that technology innovations that are aimed at economically weaker consumers need to incorporate a participative, user-engaged design process. Any applications of IoT-based innovation would take into account existing social paradigms among the target consumers viz. social acceptance [22], technological awareness [23] and consumer need [24]. This co-operative design process would also aid in the distribution of the technology among the consumers [25].

The adoption of an IoT-based innovation for the urban poor would focus not only on distribution but also on the design process. This aspect of the disruption of things is especially important as it contributes to the distribution of the system among the target consumers. In this context, variables that capture the social dynamics of the target users need to be measured. The objectives of the study are hence as follows:

To identify services that can be provided to the urban poor. To capture prevalent social dynamics that relate to technology use among the urban poor. To define a model for adoption of IoT-based innovations by the urban poor that incorporates a socially-cooperative design

process.

3. Research Methodology

The research is exploratory in nature and is conducted over four stages. In the first stage, a literature review is conducted to identify sources of IoT-based innovation for the urban poor. This is then validated by observing a family from the segment for a week and capturing their daily activities. Subsequently, we create a questionnaire based on the activity areas identified and administer the questionnaire to 20 different families in order to gauge the feasibility – technological and sociological - of IoT-based innovations in different daily processes for the segment. In the third stage, the qualitative data from the interviews is analyzed and the results are validated through a review of prior socio-economic research to arrive at the services that can be provided to the target consumers. Hence, semi-structured interviews using the data from the above analysis are conducted with third parties such as NGO workers, technology experts and volunteers. Lastly, a prototype technology system that caters to one of the processes identified in the second stage is developed and tested on respondents from the environment. The input captured in all the above stages is used to develop a model for adoption of IoT-based innovations by the urban poor.

3.1. Instruments

The literature review in the first stage considered the following seven ecosystem processes that are consistent with the UNDP India’s criteria for inclusive growth:

Housing - Includes many factors like type of construction, availability of natural light, electricity etc. Nutrition - Includes food awareness among family members, availability of food items, cooking facilities etc. Healthcare - This has factors dealing with general health, medical emergencies, and local healthcare facilities. Education - Factors like education quality, performance, scope for further training and reasons for education. Jobs - It is about the types and availability of jobs, location and preference of jobs.

201 Abhimanyu Roy et al. / Procedia Engineering 159 ( 2016 ) 199 – 209

Finances - This factor analyses financial knowledge and access to financial instruments. Communication - This component deals with collecting information, filtering it and distributing it to its users.

Next, we observed the daily activities of a family for one week. A family that has three generations was chosen to gain an

understanding of the usability of the technology for users of all ages/experience groups. After observing the daily activities of this family for duration of one week several factors were identified that were found essential for family. Based on this a questionnaire was prepared which contained 135 factors classified along line the 7 processes, as given in Table 1. The responses to these questionnaires were analyzed and used as the basis of semi-structured interviews with 12 experts.

3.2. Data Sources

The subject family in the first stage lives in Ramapir No Tekro slum in Ahmadabad, and was identified through a partner local NGO based on our specifications. The family has a 32-year old male, his wife, two children aged 9 and 6 and his mother. The slum is home to over 140,000 people, mostly migrant workers and their families from rural areas of Gujarat and Rajasthan. The other 20 validation families were randomly selected from Ramapir No Tekro.

Data was collected from the family over seven days mostly by observation but occasionally by interview. The activities of the family from the time they wake up in the morning to the time they sleep at night were observed. Their place of work and the school the kids study at were visited. All activities were recorded at the time and place.

The subject family reported the financial stand listed in Table 2. The family had taken a loan of Rs. 15000 for upgrading their house four years ago at an interest of 36% annually. They have repaid Rs. 10000 in irregular instalments. Their interest on the remaining amount comes around Rs150 per month.

No one in the family is insured for health or accidents. The father works in a factory as a lathe machine operator. His job is not permanent and he can be fired if he does not report to work for 2 months. He does not have any form of insurance. He does not come under any worker union as the unions are not prevalent in private firms in this region. His wife works as a maid along with the grandmother. She is working in 1 house and earns Rs. 600. The grandmother has been working with the household for 4 years so she is the only one who has a somewhat stable job in the family. Regarding the repayment of loan, the father gets yearly bonus at Diwali festival around the month of October every year. The bonus is equivalent to 1 month’s pay. He has been using up his bonus money every year to repay the loan with interest amount in a lump sum. Now that they have Rs. 5000 to repay they are paying interest every month. When they have a deficit they borrow a small amount from the grandmother’s employers and reduce next month’s milk expense. As the family only uses milk for tea and the children get milk from the NGO they cut down on this expense. The family took the loan from a money-lender who provides loans at a high rate of interest.

The validating responses from the other 20 families were collected from families living in the same slum. These families had five to nine members each and in all cases had three generations present.

In the third stage, we conducted interviews with twelve experts who were familiar with the proposed user base – 5 technology experts from Ahmedabad University, who have had previous experience in developing applications for socioeconomically challenged user and seven NGO volunteers/workers who have had many years of experience working with the target group of users. Based on the input from these interviews we identified five living processes which were potential sources of IoT-based innovation. We selected one of these sources – employment – and developed a prototype version of a technology system that catered to the prospective users’ employment needs – a mobile employment portal. The development process for the system utilized consultations with the experts from the third stage in order to implement a participative model of development.

Usability tests of this system were then conducted with thirteen individuals from the environment who were selected with the help of a local NGO. However, care was taken so that respondents whose technological familiarity was representative of the community’s overall technological awareness were chosen for the tests.

4. Analysis Results

4.1. Sources of innovation and potential services

Healthcare: Several healthcare factors can be improved through better awareness and access to facilities, as described in the following.

Dental care: All respondents say that they do not have regular dental checkups even though 4 of these respondents have at

least one of their family members that had a dental procedure done. In all of these cases the procedure was tooth extraction performed at a government hospital. 16 families said that no one in their families ever visited a dental clinic. Overall dental awareness is very poor and can be helped by providing necessary information through sensors that monitor their dental activities and provide requisite information regarding their needs. Ganavadiya et. al [26] has shown that spreading awareness about good dental care practices can have major improvements for underserved populations.

202 Abhimanyu Roy et al. / Procedia Engineering 159 ( 2016 ) 199 – 209

Women care: 19 out of 20 families said that no women in their family have ever visited a gynecologist. These families were both lacking the awareness about women care and also about the access to a specialized doctor. IoT sensors that monitor the content of urine can help provide women with the necessary information about their health. Bobhate and Shrivastava [27] have shown that preventive measures regarding nutrition and healthcare of women can positively impact the lives or poor urban women.

Table 1. Key factors identified for the seven processes.

Housing Education Finances Healthcare Employment Nutrition Communication -Size -Construction -Roof -Windows -Electricity -Kitchen ware -Living quarter -Mosquito nets -Furniture -Clothing -Flooring -Refrigerator -Water -Cleaning house -Toilets -Hands washing -Dental care -Hair care -Fingernails care -Women care -Baby care -Footwear use -Cleaning of utensils -Cleaning of clothes -Bathing habits

-Kids school -Frequency -Subjects studied -Study material -Language -Timing -Educational evaluation -Objectives of education -Women skills - -Current education level -Work skills -Education leaving reasons -Objectives of education -Men skills -Current education level -Work skills -Type of possible work -Education leaving reasons -Objectives of Education

- Using banks or knowledge of it -Purposes for loan -Amount of loan -Repayment -Interest rate -How to deal with late payments -Loan source -Bank authority response -Insurance - Costs (incurred or expected) -Which members of family -Hospital -Outpatient -Dental -Eye care -Insurance response -Expenses - Groceries -Electricity -Education -Transportation -Loan repayment -Cooking gas -Other (specify) -Savings (specify)

-Knowledge (which vaccination, when, to whom) -Frequency -Cost -Completion of course -Clinic - Where -How often -Needs -What clinic -What hospital -Chronic -Frequent -Long term medication -Emergencies - -How to deal with it -Previous experiences -Monitoring - How to care for family health -Special needs -Disability -Medication follow-up

-Work frequency (each capable member) -Usual income for each working member -Work location

-Number of daily meals - Each meal's contents - Cooking facilities - Types of food Shopping habits (what, when, where) -Table manners -Knowledge of healthy food -Baby food -Diet requirements -Food habits -Available choices -Seasonal variations

-Voice (phone, Skype). How many phones, phone type, cost, service provider, -VoIP -SMS -Text chat (what service provider) -Exchange photos -Exchange video -Exchange of music -Tracking, knowing where everyone is at anytime -Social activities -Physical activity

Nutrition: This can be improved through a change of user behavior.

Number of daily meals: All families said that they have only lunch and dinner, with breakfast consisting of one chapatti and a cup of tea. This choice of breakfast depends on habit more than a need. Although some family members may be employed

Table 2. Example finances of a slum family (Indian rupees)

Income

Grandmother (working as a housemaid) 1500

Father (working at a private factory) 5100

Mother (working as a housemaid) 600

Expenses

Groceries 1500

Cooking gas 1000

Electricity 800

Milk 1200

School 650

Interest on loan 150

Auto rickshaw rent for the women 1200

Cable television expense 450

Snacks and Miscellaneous 400

Deficit 150

Table 3. List of data sources

Sample Unit Family Survey

Expert Survey Usability Test

Sample Size 20 12 13

Sampling Method

Random Random Random

Technique Structured Interview

Semi-structured Interview

Survey

Total 7200 7200 7200

203 Abhimanyu Roy et al. / Procedia Engineering 159 ( 2016 ) 199 – 209

in more physically demanding professions such as construction, they ate the same breakfast. This was followed by all families. Cohen and Garrett [28] have shown that rising urban food prices negatively impact the consumption of the urban poor. An application that monitors the diet of the users and changing prices of vegetables and grains and makes that information available to the users can improve this situation.

Cooking facility: 16 out of 20 families said that they used LPG as cooking fuel, while 4 families use firewood. Out of the 16 families that use LPG, five do not have an authorized connection from any authorized agency. They buy the fuel from the black market at double the price. Firewood is a costly option as it is generally three times the price of LPG and can be very harmful to health. An application can explain the procedures and provide contact details for obtaining an authorized fuel source. This type of information may help in guiding the family to reduce their gas expense by around half. Augistina et. al [29] have shown that providing awareness and knowledge to the urban poor regarding preparation of food can improve their health considerably.

Knowledge of healthy food: 15 out of 20 families were unaware of the need of micronutrients like Iodine and Iron in the diet. They were also unaware of the food items that contain these micronutrients. An application can monitor the family’s diet and provide information about the necessity of micronutrients and the food sources they can be found in. Ahmed et. al [30] showed that providing knowledge about the nutritional benefits of certain foods can improved the health of urban slum children in Bangladesh.

Employment for the urban poor can be improved through better awareness and access to job opportunities. 40% of individual adult respondents were semi-skilled yet unemployed. Furthermore, 25% of the female respondents had attained professional skills such as computer proficiency but were employed in jobs such as maids and sweepers. Agarwal [31] demonstrated the skill gap in the informal market in India and a system that connects job candidates with employers coupled with an online and connected education system can rectify this situation.

Education: Quality of education for the urban poor can be improved significantly through access to better education materials and resources.

Choice of school: 12 out of 41 children go to a private school to study. The fees for a private school is generally 2-3 times

higher than the public schools. Vijay [32] noted the increasing expenditure on children’s education among the urban poor. A system that provides access massively online open courses could improve the quality of education for the target users.

Study Material: 50% of the families had bought study material for their children, while 15% had received it through donations. This corresponds with previous research that showed that parents had been purchasing study material which increased their economic burden [33]. Applications can provide high quality and more updated study material to students in a cost-effective manner.

Education Performance: 11 out of the 20 families reported that their children had displayed poor performance in school. When asked specifically about subjects such as mathematics and science, 80% of the families reported poor performance. This is similar to previous research that showed that education of slum children remained poor despite regularly attending school [34]. This situation can be improved considerably through applications that provide access to good quality study material and sensors that monitor the students’ activities and suggest necessary changes as required.

Finances: The urban poor can have better financial security through access to basic savings and loans instruments and

knowledge about the due diligence process of banks.

Loans: Four out of 20 families have approached banks for a loan. All of the families were denied loans without any specific reasoning. Nine out of 20 families have taken a loan for different purposes. There are different sources of loans available e.g. banks, micro-finance institutions (MFIs) and local moneylenders. Two of the families who have taken a loan from local moneylenders do not have a bank account. Even though people have a bank account, they still prefer to take a loan from other sources even if their interest rates are higher than banks. Banks give loans to slum families typically at rate of 25% per annum which is high as it is a risky segment but is still better option in terms of interest rate as MFIs charge around 36-50% and local moneylenders can charge anything from 60% to 600%. These high interest rates were also found by prior research [36]. Awareness regarding procedures and documentation required for taking a loan from a bank given through applications will be very helpful in securing better loans.

4.2. Usability

Out the five identified sources of innovation for the urban poor, we picked one source – employment – and constructed a system to cater to it. The system consisted of a registration part for new job candidates and a notification part that could (i) broadcast alerts every time a new job was posted, and (ii) an SMS notification system which would alert suitable candidates about potential job opportunities. The notification system was constructed on Awaaz.de, a broadcast and interactive voice service available in Ahmedabad. The messages were recorded in Gujarati as that was the local language. After construction of the system, it was tested with 13 individuals. All of the members had their tests video recorded and their feedback taken. It was on

204 Abhimanyu Roy et al. / Procedia Engineering 159 ( 2016 ) 199 – 209

this feedback that the analysis is conducted. Out of the 13 testers, 9 were women and 4 were men. All the testers found that the sound, voice, words and language of all the messages were very clear and understandable. All 13 testers thought that the system was very easy to follow and did not require any special technical ability to operate. They all thought that they could operate the system on their own and could explain it to family and friends who may be interested in using the system later on. Four respondents made mistakes while using the system. However, these mistakes were not easily reversible and required reregistration to complete the process. None of the testers who committed errors abandoned the use of the system because of the errors. All the testers were confident that they registered on the system and they felt that they would find a job by providing the information required in the registration system. All testers would recommend the use of the system. However, three of them wanted some training to be given before the first use of the system.

5. Literature Examples of the Sources of Innovation

In the previous section, we identified 5 sources of disruptive innovation. Although IoT remains at a relatively nascent state, there are a number of innovations and applications that are currently in development that cater to these sources. In this section, we explore some of these technologies.

5.1. Healthcare

IoT can significantly reduce costs and increase efficiency to provide better healthcare to the urban poor. These applications would be able to identify and authentic patients, automatically collect healthcare relate data and predict outcomes based on the analysis of the collected data and provide access to preventive healthcare-related information [37]. Jovanov and Milenkovic [38] found uses of wearable devices in providing preventive information about diseases such as diabetes to patients. Khatoon, Hill and Walmsley [39] found that a combination of IoT, cloud storage and 3D-printing was effective in providing dental-care services in remote locations. Aggregate data collected from personal wearable devices can also be used to predict the occurrence of seasonal diseases such as influenza and raise awareness about the benefits of preventive vaccinations for these diseases among the general public in a timely manner [40]. Paschou et. al [41] found that IoT devices can greatly reduce costs for standard medical tests such as blood tests and urine sampling. These testing methods combined with the integration of data into medical databases enabled by IoT allows for faster diagnosis of diseases.

5.2. Nutrition

There are major benefits of monitoring dietary intake information. Individuals would be able to understand their dietary intake and improve their health significantly. Nutritionists primarily use manual self-reporting processes such as food journals to cover personal diets. Steele [42] analyzed sensor-based approaches to capture dietary information automatically and found that the process was far more efficient in capturing information accurately. Wei et. al [43] found that informatics can guide people in their nutrition intake which prevent them from insufficient and/surplus intake which prevents a negative impact on health. Boulos et. al [44] showed that automated food and drink recognition techniques that connected to cloud-based databases could match the near-infrared spectra of food and drink can be used to recognize food items. These recognition techniques combined with automated measurement of portion sizes and ingredient nutrition information would enable devices to associate nutrition information with specific foods and guide users to appropriate diet items.

5.3. Employment

A report in 2013 from consulting firm McKinsey and Co. [45] elaborated on the use of IoT in manpower planning and skilling up employees. Monitoring of trends in production and overall market conditions and analyzing the same in the context of historical data, companies can plan hiring activities well in advance and employees can also be skilled appropriately for future challenges. This has a significant impact for the urban poor especially as a large portion of employment for the urban poor is in the informal economy which is characterized by seasonal employment [46].

5.4. Finances

The sea change that IoT can bring in the management of personal finances can be attributed to the analysis of historical spending data of others and mapping the same to an individual’s current set of financial options. Information Technology firm Accenture predicted in 2015 the growth of living services enabled by IoT [47]. As an example, let’s say an individual is faced with an upcoming financial decision such as funding the cost of surgery. An IoT system can take account of the individual’s current physiological state through sensors and find a set of suitable hospitals and treatments for them. Next, the system would be able to gauge the future cost of surgery on the basis of the current cost of surgery and historic rates of increase or decrease in costs. This future cost of surgery for different hospitals and treatments would then be compared to the historical savings rate data for the individual and the rate of return on the investments (bank deposits) they have made. Based on this comparison a final set

205 Abhimanyu Roy et al. / Procedia Engineering 159 ( 2016 ) 199 – 209

of affordable hospitals and treatments would be determined. After the individual chooses one or more of these options, the system would be able to supply them information regarding the savings that they would require to fund their surgery and the investments (in terms of the rate of return) that they would have to make in order to achieve the same. Similarly, the optimal cost for servicing appliances or education or any other financial decision can also be reached.

5.5. Education

IoT can severely bring down costs for education. IoT can extend education resources to beyond the classroom and make available quality educational material at a fraction of the cost of existing channels. Zhiqiang and Junming [48] found that the key to effective integration of IoT in education lay in the development of educational middleware that incorporated students’ progressive evaluation and existing teaching platforms. Tagging digitized versions of textbooks and making the same available to students produced gains in academic performance. In the book ‘The Zero Marginal Cost Society’, Jeremy Rifkin [49] explores the dramatic decrease in the cost of education that is brought about by Massive Open Online Courses (MOOCs). Increasing the accessibility to MOOCs for the general public after identifying relevant skill sets that they could benefit by learning would produce a cohort of professionals who are industry ready.

6. Innovations Adoption Model

In our research, to define a model for adoption of IoT-based applications by the urban poor, we first began by finding possible sources for such innovations. These innovations would be sparked by consumer need, hence we began with 7 processes that corresponded with the UNDP criteria for inclusive growth – housing, nutrition, healthcare, employment, finance, education and communication. Five of these seven – nutrition, healthcare, nutrition, employment and education – were found to be suitable for IoT-based innovations. Most of the factors that affected housing such as ventilation and construction were not suitable for IoT-based interventions. As for communication, while mobile phones do house sensors such as accelerometers that are integral to IoT, there applications can be found in other avenues such as healthcare and cell phones adequately fulfill the communication aspect for the respondents.

Once an innovation that caters to one of these sources has been determined, we move to the design phase. Design has two components – technology and social integration. Technology refers to affordability and prevalence of technology among the target users. In our case, we found that all 20 of the families that we had interviewed had access to mobile phones and hence they were both affordable and widely available.

The second component – social integration refers to the cultural integration of the technology and its general learnability. Cultural integration refers to what extent the system is familiar to the potential users. In our case, an interactive voice response system is used by the Indian government for booking of cooking gas. Hence, the users were generally familiar with using the technology. Learnability, on the other hand, focuses on the difficulty or ease experienced by the user in using the system. Since, our users were familiar with using an interactive voice response system, they could quite effectively use it to apply for jobs.

Once the system is finally ready to be implemented, it must give its users excellent service. ‘Service Excellence’ comes about as the result of three parameters – support, training and benefits. Several of the testers made errors during the usability tests and hence required our support to rectify them. If a technology-enabled service wants to gain mass-adoption among the urban poor, it would need to have a good support team in place to handle and resolve user errors. The lack of this feature would directly affect the usability component of awareness and may lead to negative word-of-mouth/publicity by users. Training refers to proper instructions that are provided to the users to enable system usage. Errors and system usage issues that may crop up need to be minimized. As stated previously, errors occurred in 50% of the usability tests that were conducted and 30% of the testers wanted more training to be provided.

Benefits refers to the perceived value of the system for the users. These benefits can be in terms of cost and/or time. In our case, the users perceived the system as saving them time in finding a job. However, since jobs also have a financial component – the system’s use also has a cost benefit. Despite experiencing problems during registration, all thirteen respondents in our usability test said that they would recommend the system to their friends and family. This was because they perceived some value in use of the system.

It must be pointed out, at this juncture, that we decided on this development after consultations with experts who were familiar with the sociology and technological awareness of the target users. This coupled with the willingness of the users to recommend the system, necessitates the use of a participative design process for IoT-based innovations for the urban poor.

Service Excellence if implemented correctly, would lead to a satisfied user. These satisfied users would then be used to increase adoption of the system among the urban poor. For this, three channels can be used – word of mouth, advertising and social media. Word-of-mouth refers to recommending the system to friends and family. In the usability tests we found that participants would be more willing to use the system if a friend had recommended it. Hence, positive word-of-mouth has a pivotal role to play in system adoption. We also found that families tended to use products (soaps, food items etc.) and brands that were widely advertised. We extrapolate that this would be the same for an IoT innovation, i.e., advertising with positive endorsements from satisfied users will bring in increased adoption. The final channel is social media. In the communication portion of the survey we noted the increasing prevalence of social media such as Facebook and Whatsapp in the lives of the

206 Abhimanyu Roy et al. / Procedia Engineering 159 ( 2016 ) 199 – 209

urban poor. Although, only three of the families had members who used these platforms, they had started using it fairly recently – within the past year. User-generated content on these platforms would have a major impact on the adoption of IoT innovations by the urban poor. Photos, videos and messages shared by the users that referred to their positive experience while using the innovation would spur its adoption very quickly – in a manner similar to a viral post.

Fig. 1. A model to facilitate adoption of IoT-based innovations by the urban poor.

7. Example Implemented Solutions

Business process models were constructed for the identified processes, e.g. Fig. 2 depicting a health management diagram. A realization of a technology system for job management is depicted in Fig. 3. The health care and employment management solutions were deployed within Microsoft Azure cloud infrastructure and technical details are reported elsewhere, but the design considered the social mechanisms prevalent among the urban poor as well as feasible technological tools. Due to the limited financial resources of the target customers, affordability and access to technology would be hindered and any design process would need to incorporate this. In our study, we chose feature cell phones as they were widely available. Similarly, social acceptance of the technology had to be accounted for. We designed an interactive voice response system as the consumers were familiar with the use of such as system for e.g. booking cooking gas from government agencies.

207 Abhimanyu Roy et al. / Procedia Engineering 159 ( 2016 ) 199 – 209

Fig. 2. Health management: user’s communications with providers (clinics and hospitals).

Fig. 3. Job management realization for the slum dwellers

8. Study Limitations and On-going Work

The studies reported in this paper are based on quantitative and qualitative data captured from urban slum communities in Ahmedabad and Kolkata, India. The analysis of captured data as well as relevant literature create a foundation model for implementing actual solutions of IoT ideas for the poor. The limited empirical data and the on-going implementations and investigations makes this paper broad in scope. However, the reported model and pilots emphasize the importance of this approach that could have significant impact if fully realized.

Further data analysis is ongoing with wider scope in-depth qualitative analysis based on grounded theory, and the impact of the piloted solutions on the community is being evaluated. The cloud-based Microsoft Azure platform is proving an excellent infrastructure for experimenting with the adoption model.

208 Abhimanyu Roy et al. / Procedia Engineering 159 ( 2016 ) 199 – 209

9. Conclusions

Consumer needs govern the sources of IoT-based innovation for the urban poor. We have identified 5 sources for such innovations – nutrition, healthcare, employment, education and/or finances. All of these sources are in accordance with the UNDP criteria for inclusive growth.

An implemented system would need to provide quality service to the users. The users need to experience tangible benefits through the use of the system. The issues they experience would need to be handled with a support system. First time users would also need to be given training in order to help them seamlessly use the system. Excellent service provided to consumers would lead to satisfied users.

The positive endorsements of these satisfied users would be used in three channels to increase adoption of the IoT-based innovation. These channels include direct advertising to the target users, word-of-mouth publicity that drives more users to the solution and viral posts on social media, which our research shows as an emerging medium to reach the urban poor.

References

[1] International Telecommunication Union. (June 15, 2012). Internet of Things Global Standards Initiative. In International Telecommunication Union. Retrieved January 21, 2016, from http://www.itu.int/en/ITU-T/gsi/iot/Pages/default.aspx.

[2] Tan, L., & Wang, N. (2010, August). Future internet: The internet of things. InAdvanced Computer Theory and Engineering (ICACTE), 2010 3rd International Conference on (Vol. 5, pp. V5-376). IEEE.

[3] Ashton, K. (2009). That ‘internet of things’ thing. RFiD Journal, 22(7), 97-114. [4] Atzori, L., Iera, A., Morabito, G., & Nitti, M. (2012). The social internet of things (siot)–when social networks meet the internet of things: Concept,

architecture and network characterization. Computer Networks, 56(16), 3594-3608. [5] Schaffers, H., Komninos, N., Pallot, M., Trousse, B., Nilsson, M., & Oliveira, A. (2011). Smart Cities and the Future Internet: Towards Cooperation

Frameworks for Open Innovation. Future Internet Assembly, 6656, 431-446 [6] Istepanian, R. S. H., Hu, S., Philip, N. Y., & Sungoor, A. (2011, August). The potential of Internet of m-health Things “m-IoT” for non-invasive glucose level

sensing. In Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of the IEEE (pp. 5264-5266). IEEE. [7] Sanchez, L., Galache, J. A., Gutierrez, V., Hernandez, J. M., Bernat, J., Gluhak, A., & Garcia, T. (2011, June). Smartsantander: The meeting point between

future internet research and experimentation and the smart cities. InFuture Network & Mobile Summit (FutureNetw), 2011 (pp. 1-8). IEEE. [8] Coughlan, T., Brown, M., Mortier, R., Houghton, R. J., Goulden, M., & Lawson, G. (2012, November). Exploring Acceptance and Consequences of the

Internet of Things in the Home. In Green Computing and Communications (GreenCom), 2012 IEEE International Conference on (pp. 148-155). IEEE. [9] Ravallion, M., & Chen, S. (2007). China's (uneven) progress against poverty.Journal of development economics, 82(1), 1-42. [10] Duflo, E., Galiani, S., & Mobarak, M. (2012). Improving Access to Urban Services for the Poor. [11] Boettiger, S. (2014). Tempered Enthusiasm for Digitally Enabled Networks in International Development. innovations, 9(1-2), 125-150. [12] Kortuem, G., Bandara, A. K., Smith, N., Richards, M., & Petre, M. (2013). Educating the Internet-of-Things generation. Computer, 46(2), 53-61. [13] Dlodlo, N. (2013). Potential applications of the internet of things technologies for South Africa’s health services. In International Conference on ICT for

Africa (pp. 20-23). [14] Rohokale, V. M., Prasad, N. R., & Prasad, R. (2011, February). A cooperative Internet of Things (IoT) for rural healthcare monitoring and control.

In Wireless Communication, Vehicular Technology, Information Theory and Aerospace & Electronic Systems Technology (Wireless VITAE), 2011 2nd International Conference on (pp. 1-6). IEEE.

[15] Musiyandaka, D., Ranga, G., & Kiwa, J. F. (2013). An analysis of factors influencing success of ICT4D projects: a case study of the Schools Computerisation Programme in Mashonaland West Province, Zimbabwe. The Journal of Community Informatics, 9(4).

[16] Ginsburg, O. M., Chowdhury, M., Wu, W., Chowdhury, M. T. I., Pal, B. C., Hasan, R., ... & Mahmud, S. (2014). An mHealth model to increase clinic attendance for breast symptoms in rural Bangladesh: can bridging the digital divide help close the cancer divide?. The oncologist, 19(2), 177-185.

[17] Free, C., Phillips, G., Galli, L., Watson, L., Felix, L., Edwards, P., ... & Haines, A. (2013). The effectiveness of mobile-health technology-based health behaviour change or disease management interventions for health care consumers: a systematic review. PLoS Med, 10(1), e1001362.

[18] Lee, M. K., & Turban, E. (2001). A trust model for consumer internet shopping. International Journal of electronic commerce, 6(1), 75-91. [19] Sathye, M., & Beal, D. (2001). Adoption of electronic commerce by SMEs: Australian evidence. Journal of E-Business, 1(1), 1-11. [20] Kim, H. W., Chan, H. C., & Gupta, S. (2007). Value-based adoption of mobile internet: an empirical investigation. Decision Support Systems, 43(1), 111-

126. [21] Heeks, R. (2008). ICT4D 2.0: The next phase of applying ICT for international development. Computer, 41(6), 26-33. [22] Ramachandran, D., Kam, M., Chiu, J., Canny, J., & Frankel, J. F. (2007, April). Social dynamics of early stage co-design in developing regions.

InProceedings of the SIGCHI conference on human factors in computing systems (pp. 1087-1096). ACM. [23] Lal Dey, B., Binsardi, B., Prendergast, R., & Saren, M. (2013). A qualitative enquiry into the appropriation of mobile telephony at the bottom of the

pyramid. International Marketing Review, 30(4), 297-322. [24] Ramani, S. V., SadreGhazi, S., & Duysters, G. (2012). On the diffusion of toilets as bottom of the pyramid innovation: Lessons from sanitation

entrepreneurs. Technological Forecasting and Social Change, 79(4), 676-687. [25] Vachani, S., & Smith, N. C. (2007). Socially responsible distribution: distribution strategies for reaching the bottom of the pyramid. INSEAD Business

School Research Paper, (2008/21). [26] Ganavadiya, R., Chandrashekar, B. R., Goel, P., Hongal, S. G., & Jain, M. (2015). Mobile and portable dental services catering to the basic oral health needs

of the underserved population in developing countries: a proposed model. Annals of medical and health sciences research, 4(3), 293-304. [27] Bobhate, P., & Shrivastava, S. (2011). A cross sectional study of knowledge and practices about reproductive health among female adolescents in an urban

slum of Mumbai. Journal of Family and Reproductive Health, 5(4), 117-124. [28] Cohen, M. J., & Garrett, J. L. (2010). The food price crisis and urban food (in) security. Environment and Urbanization, 22(2), 467-482. [29] Agustina, R., Sari, T. P., Satroamidjojo, S., Bovee-Oudenhoven, I. M., Feskens, E. J., & Kok, F. J. (2013). Association of food-hygiene practices and

diarrhea prevalence among Indonesian young children from low socioeconomic urban areas. BMC public health, 13(1), 977. [30] Ahmed, F., Rahman, A., Noor, A. N., Akhtaruzzaman, M., & Hughes, R. (2006). Anaemia and vitamin A status among adolescent schoolboys in Dhaka

City, Bangladesh. Public health nutrition, 9(03), 345-350.

209 Abhimanyu Roy et al. / Procedia Engineering 159 ( 2016 ) 199 – 209

[31] Agrawal, T. (2012). Vocational education and training in India: challenges, status and labour market outcomes. Journal of Vocational Education & Training, 64(4), 453-474.

[32] Vijay, D. (2013). Structure and Mobility in a Delhi Slum: 1988–2013. Social Change, 43(4), 595-615. [33] Tsujita, Y. (2013). Factors that prevent children from gaining access to schooling: A study of Delhi slum households. International Journal of Educational

Development, 33(4), 348-357. [34] Chakraborty, S. (2014). Reaching the un-reached: Case studies of the learning initiatives of the street children in Kolkata. [35] Mahajan, S. S., & Kalel, N. (2013). An Assessment of Potential Financial Inclusion of Slum Dwellers Through Business Correspondent Model. IUP Journal

of Bank Management, 12(4), 7. [36] Dehejia, R., Montgomery, H., & Morduch, J. (2012). Do interest rates matter? Credit demand in the Dhaka slums. Journal of Development Economics,

97(2), 437-449. [37] Gubbi, J., Buyya, R., Marusic, S., & Palaniswami, M. (2013). Internet of Things (IoT): A vision, architectural elements, and future directions. Future

Generation Computer Systems, 29(7), 1645-1660. [38] Jovanov, E., & Milenkovic, A. (2011). Body area networks for ubiquitous healthcare applications: opportunities and challenges. Journal of medical systems,

35(5), 1245-1254. [39] Khatoon, B., Hill, K. B., & Walmsley, A. D. (2013). Can we learn, teach and practise dentistry anywhere, anytime?. British dental journal, 215(7), 345-347. [40] Liyanage, H., de Lusignan, S., Liaw, S. T., Kuziemsky, C., Mold, F., Krause, P., ... & Jones, S. (2014). Big Data Usage Patterns in the Health Care Domain:

A Use Case Driven Approach Applied to the Assessment of Vaccination Benefits and Risks: Contribution of the IMIA Primary Healthcare Working Group. Yearbook of medical informatics, 9(1), 27.

[41] Paschou, M., Sakkopoulos, E., Sourla, E., & Tsakalidis, A. (2013). Health Internet of Things: Metrics and methods for efficient data transfer. Simulation Modelling Practice and Theory, 34, 186-199.

[42] Steele, R. (2014). An overview of the state of the art of automated capture of dietary intake information. Critical reviews in food science and nutrition, (just-accepted), 00-00.

[43] Wei, M. H., Yang, L. Q., Zhou, Q. L., & Zheng, J. G. (2013, October). Construction of Optimal Allocation System on the Edible Agricultural Products Components of the Nutrition and Health Care Based on Internet of Things. In Applied Mechanics and Materials (Vol. 347, pp. 3194-3198).

[44] Boulos, M. N. K., Yassine, A., Shirmohammadi, S., Namahoot, C. S., & Brückner, M. (2015). Towards an “Internet of Food”: Food Ontologies for the Internet of Things. Future Internet, 7(4), 372-392.

[45] Manyika, J., Chui, M., Bughin, J., Dobbs, R., Bisson, P., & Marrs, A. (2013). Disruptive technologies: Advances that will transform life, business, and the global economy (Vol. 180). San Francisco, CA, USA: McKinsey Global Institute.

[46] McGahan, A. M. (2012). Challenges of the informal economy for the field of management. The Academy of Management Perspectives, 26(3), 12-21. [47] Accenture. (2015). The Era of Living Services. In Accenture. Retrieved February 5, 2016, from https://www.accenture.com/us-en/insight-living-services-

from-accenture-digital.aspx. [48] Zhiqiang, H., & Junming, Z. (2011). The Application of Internet of Things in Education and Its Trend of Development [J] Modern Distance Education

Research, 2, 019. [49] Rifkin, J. (2014). The zero marginal cost society: The internet of things, the collaborative commons, and the eclipse of capitalism. Palgrave Macmillan.