IFS741 SYSTEM DESIGN

DOCUMENT TEMPLATE

STAYCLOSE

Date of Initial review:

PROJECT:
TABLE OF CONTENTS

1 INTRODUCTION................................................................................................................4

1.1 PURPOSE OF THIS DOCUMENT...................................................................................4

1.2 AUDIENCE............................................................................................................. 4
1.3 EXECUTIVE SUMMARY.............................................................................................5
1.3.1 SOLUTION PROPOSAL.......................................................................................5
1.3.2 CONCEPT DESCRIPTION AND MOTIVATION......................................................6
1.3.3 POTENTIAL VALUE...........................................................................................6
1.3.4 RELEVANCE....................................................................................................7
1.4 STANDARDS OF DESIGN...........................................................................................7
1.5 SIMILAR SOLUTIONS..........................................................................................8
1.6 STAKEHOLDERS: ROLES AND RESPONSIBILITIES..............................................................9
1.6.1 STAKEHOLDERS & ROLES..................................................................................9
2 DESIGN GUIDELINES AND APPROACH.................................................................................10

2.1 GENERAL OVERVIEW.............................................................................................10

2.2 GOALS, GUIDELINES, AND RISKS............................................................................12
2.2.1 SECURITY.....................................................................................................12
2.2.2 PRIVACY......................................................................................................12
2.2.3 LAWS OF IDENTITY PRIVACY............................................................................12
2.2.4 POPIA........................................................................................................13
2.2.5 TRUST & (OR VS) TRANSPARENCY....................................................................15
2.2.6 AREAS OF RISK............................................................................................ 16
3 REQUIREMENTS & DESIGN...............................................................................................18

3.1 GENERAL OVERIEW...............................................................................................18

3.2 BUSINESS REQUIREMENTS......................................................................................19
3.3 FUNCTIIONAL REQUIREMENTS (FOR BOTH SIDES OF PLATFORM)...........................19
3.4 ASSUMPTIONS/CONSTRAINTS/DEPENDENCIES/RISKS...................................................21
3.5 USER SCENARIOS..............................................................................................22
4 ARCHITECTURE DESIGN................................................................................................... 23

4.1 SYSTEM ARCHITECTURE..........................................................................................23

4.2 HARDWARE ARCHITECTURE.....................................................................................25

2 | 39
 4.2.1 CONNECTIVITY..............................................................................................26
4.3 SOFTWARE ARCHITECTURE......................................................................................26
4.4 DATA & INTERNAL COMMUNICATION.......................................................................29
5 SOLUTION DESIGN..........................................................................................................31

5.1 VISUAL DESIGN.....................................................................................................31

5.1.1 WIREFRAMES & VISUALS................................................................................31
5.2 DATABASE DESIGN................................................................................................35
5.3 DATA CONVERSION AND DATA ANALYTICS...........................................................35
6 EXTERNAL INTERFACE DESIGN.......................................................................................... 36

6.1 GOOGLE MAPS.................................................................................................... 36

6.2 MEDICAL EQUIPMENT........................................................................................... 36
6.2.1 HEART RATE & BLOOD PRESSURE MONITOR......................................................36
6.2.2 TEMPERATURE MONITOR...............................................................................36
6.3 GENERAL OFFICE EQUIPMENT.................................................................................36
6.3.1 PRINTER/SCANNER........................................................................................36
7 LIVING LABS DEMONSTRATION AND FEEDBACK.......................................................37

7.1 TEST GROUPS...................................................................................................... 37

7.1.1 TEST GROUP 1: HEALTHCARE PROFESSIONALS....................................................37
7.1.2 TEST GROUP 2: CUSTOMER/PATIENT................................................................38
8 APPENDIX.........................................................................ERROR! BOOKMARK NOT DEFINED.

LIST OF FIGURES

3 | 39
 1 INTRODUCTION

1.1 PURPOSE OF THIS DOCUMENT

The end goal for the IFS741 course is to use Design Science Research (DSR) as a framework to guide
the creation of innovative platforms, such as websites or apps. For this assignment, I have decided to
concentrate on the fourth topic, AI Caregivers: The Future of Elderly Care. investigating how AI might
be used to deliver efficient and friendly healthcare. The design process is organized according to the
tried-and-true methodology described by Peffers et al. (2007), whose six iterative processes for
presenting and conducting DSR have grown to become a standard in the information systems
industry.

The goal is to provide a thorough and rigorous platform that not only satisfies the technical and
ethical standards but also withstands criticism from the information systems literature and other
pertinent academic domains by aligning our design approach with the DSR methodology. I will work
through the following phases: problem identification and motivation; goal setting for the solution;
design and development; evaluation; user input for demonstration; and, lastly, communication of our
conclusions and design outcomes. With privacy, usability, and user experience at its core, this
methodical approach will guarantee that our AI caregiver platform contributes significantly to the
rapidly developing field of AI in caring.

1.2 AUDIENCE

The wandering prevention software caters to a number of important demographics. The app's real-
time location monitoring and alert features are intended to help primary caregivers, such as family
members, home aids, or professional caregivers, since they offer peace of mind by guaranteeing the
safety of their loved ones. Based on the information gathered, medical professionals—such as
physicians, nurses, and specialists—can utilize the app to proactively manage patients' ailments by
keeping an eye on behaviour patterns. The software may be used by senior care facilities, such as
assisted living facilities or nursing homes, to improve resident safety, lessen the need for continual
physical supervision, and provide better care overall.

Additionally, the program allows senior users to travel freely within safe zones while being watched.
These users have moderate cognitive impairment or early-stage dementia and want to keep some

4 | 39
degree of freedom. In addition, the app can help emergency responders and public safety by giving
crucial location information to elders who may wander out, allowing for prompt interventions and
recovery. Finally, the app could be helpful to health insurance companies in providing or suggesting
an affordable solution that improves client safety and lowers the risks and expenses related to
emergency medical procedures.

1.3 EXECUTIVE SUMMARY

The suggested remedy, called roaming prevention, is an AI-powered platform that tracks the
whereabouts of elderly people suffering from dementia and prevents them from roaming, improving
their safety and wellbeing. The program uses artificial intelligence (AI) monitoring tools to track real-
time location data and make sure elders don't leave defined safe zones without alerting caretakers
right away. This gives older people a better sense of security and gives family members and
caregivers peace of mind.

The app fills a vital gap in society, especially as the world's population ages and cognitive disorders
like dementia become more common. Improved senior autonomy, proactive healthcare management
via data analysis, and affordable monitoring options are the main advantages. To provide a flexible
and intuitive experience, the app combines behavioural data analysis, local networks, and geofences
that may be customized.

The platform's integrated privacy safeguards guarantee adherence to data protection laws such as
POPIA. Additionally, by emphasizing real-time data exchange and AI-assisted decision-making, it
lessens the workload on healthcare institutions. This initiative is important because it promotes
coordinated care between families, healthcare professionals, and caregivers, hence supporting senior
safety and contemporary elder care. Using the Design Science Research (DSR) approach, the solution
is meticulously built to meet technological, ethical, and usability requirements.

1.3.1 SOLUTION PROPOSAL

Request Title Wandering Prevention App for elderly care

Customer Contact Sarah Mitchell
Project Manager
555-123-4567

5 | 39
 [email protected]
Document Version
Jason Thompson
Lead Designer - User Interface
Designer Contact
555-234-5678
[email protected]

1.3.2 CONCEPT DESCRIPTION AND MOTIVATION

1.3.2.1 APPLICATION IDEA:

Wandering Prevention: Implement AI tracking systems to monitor the movements of seniors with
dementia, ensuring they do not wander off and get lost.

● As the global population ages, ensuring the safety and well-being of the elderly, particularly
those with cognitive impairments like dementia, has become a critical concern. A key issue for
this group is the risk of getting lost or wandering. Developing an app to prevent such incidents
offers numerous benefits, including enhanced security, peace of mind for caregivers, and
support for the elderly’s independence.

● The app's primary advantage is improving safety by tracking the real-time location of seniors,
alerting caregivers when they leave a designated safe zone, and allowing for quick
intervention in potentially dangerous situations. This not only protects the elderly but also
relieves caregivers of the constant worry about their loved one's safety. Moreover, the app
promotes autonomy by allowing seniors to move freely within set boundaries, balancing
independence with security. Additionally, it can collect valuable health data, identifying
behavioural changes that may indicate cognitive decline, thereby enabling proactive
healthcare management. With customizable alerts and geofences, the app is adaptable to
individual needs and offers a cost-effective alternative to other monitoring methods.
Integration with local networks further strengthens its utility, making it a vital tool in modern
senior care.

1.3.3 POTENTIAL VALUE

6 | 39
Based on the network theory of power, this elderly safety app has the potential to generate several
forms of power and value. It can exercise network-making power by establishing the rules and
protocols for the network connecting elderly users, caregivers, and healthcare providers (Castells,
2011). The app may also wield networking power through inclusion or exclusion of users or features.
Within the network, certain nodes like caregivers may have more influence, creating networked
power dynamics. The app generates significant social value by enhancing elderly safety and providing
peace of mind for caregivers, while also supporting seniors' independence. It can create health data
value through collecting behavioural data for initiative-taking healthcare management. Economically,
it offers a cost-effective monitoring solution. If successful in connecting different networks (elderly
care, healthcare, family), it could exercise switching power. The app also has potential to shape
cultural attitudes towards aging and care, generating cultural power. Its overall power and value will
depend on adoption rates, effective implementation, and how it addresses challenges like privacy
concerns and ethical considerations in elderly care.

1.3.4 RELEVANCE

This network is highly relevant for several reasons. Firstly, it addresses a critical societal need by
focusing on the safety and well-being of elderly individuals, particularly those with cognitive
impairments, in the context of an aging global population. Secondly, it leverages modern
communication and tracking technologies to solve real-world problems in healthcare and elder care,
demonstrating the practical application of technology for social benefit. The network also balances
the autonomy of seniors with their safety, tackling a complex challenge in elder care by providing
support while ensuring security. Additionally, by collecting and analysing behavioural data, it
contributes to initiative-taking health management and early detection of cognitive decline, aligning
with the trend of data-driven healthcare. The network interconnects various stakeholders—elderly
individuals, caregivers, and healthcare providers—improving coordination and communication. It also
illustrates how different forms of network power operate in a real-world context, including network-
making, networking, and networked power. The ethical considerations surrounding privacy, consent,
and technology use in vulnerable populations are also addressed, making it relevant to ongoing
discussions about digital age ethics. Finally, as a cost-effective solution, it contributes to managing
healthcare costs for aging populations, reflecting broader societal trends and technological
advancements.

1.4 STANDARDS OF DESIGN

7 | 39
We will use the four-cycle approach of design science research to assure rigor in creating an app for
senior safety, especially for those with cognitive impairments. The software will be grounded on
current theories on cognitive decline, elder care, and mobile health applications thanks to the rigor
cycle, which will also add fresh perspectives to the body of knowledge. Iterative prototyping and
senior- and caregiver-specific usability testing are part of the design cycle. Stakeholders will be
involved in the relevant cycle to better understand their demands and conduct field tests in actual
environments. The cycle of change and impact will track the app's wider implications on senior care
practices and adjust to changes in regulations and technology.

We will design for both immediate utility and long-term adaptability across these cycles, strike a
balance between detailed literature reviews and rapid prototyping, and conduct rigorous evaluations
utilizing a combination of methodologies. It will be crucial to take ethics into account, especially
about consent and privacy. Multidisciplinary cooperation will guarantee an all-encompassing
strategy. Rigor will be further enhanced by thorough documenting of the design process and frequent
reflection on how the app's development supports or contradicts accepted beliefs. With this
technique, an app will be produced that is theoretically sound, useful in real life, and adds significant
insights to the area.

1.5 SIMILAR SOLUTIONS | MARKET ANALYSIS

There are several networks and solutions out there that accomplish comparable goals to the app.
These mostly consist of wearable technology, GPS trackers, and specific apps made to track the
whereabouts of senior citizens, especially those suffering from cognitive impairments like dementia
or Alzheimer's disease.

For example, GPS monitoring gadgets like Angel Sense and GPS Smart Sole made especially for those
with cognitive impairments. These gadgets are equipped with capabilities like geofencing, real-time
location tracking, and alarms that notify caregivers when the user departs from a specified safe
region. Like this, apps like Safe365 and Trackimo offer location services, emergency alerts, and
occasionally health monitoring tools as part of their mobile-based solutions for tracking and
monitoring the elderly.

The app StayClose might set itself apart, though, by providing improved features like sophisticated
data analytics for initiative-taking healthcare management, smooth connection with neighbourhood
networks, and extensive user customization to suit specific requirements. This feature set may fill up
the gaps in current offerings, providing a more complete aged care solution.

8 | 39
 1.6 STAKEHOLDERS: ROLES AND RESPONSIBILITIES

1.6.1 STAKEHOLDERS & ROLES

The following table identifies the product stakeholders who are instrumental in strategic, tactical, and
operational decision making, and roles involved in this phase of the project, including components of
system, software, and architecture design. The contact details are for the express use of the project
manager, and other designated roles to effect calling of meetings, distribution of tasks and feedback,
and solicitation of information.

Name Role Phone Email

Sarah Mitchell Project Manager 555-123-4567 [email protected]

Jason Thompson Lead Designer, User Interface 555-234-5678 [email protected]

Robert Johnson System Architect 555-234-5678 [email protected]

Emily Davis Software Developer 555-456-7890 [email protected]

Daniel White Quality Assurance Lead 555-567-8901 [email protected]

Lisa Carter Business Product Owner 555-678-9012 lisa.carter@gmail.

Table 1-Project Roles

1.7. ORGANISATIONAL AND SOCIAL IMPACT OF PROPOSED NEW PLATFORM

The introduction of the wandering prevention platform offers transformative organizational and
social benefits, addressing pressing challenges associated with elder care and dementia
management. As societies age and cognitive disorders become more prevalent, this platform
presents an innovative solution that enhances safety, autonomy, and overall quality of life for elderly
individuals while supporting caregivers, families, and healthcare institutions.

On an organizational level, the platform reduces the workload for healthcare providers by
automating routine monitoring tasks. With real-time geofencing alerts and health tracking
capabilities, caregivers and medical professionals can focus on more critical aspects of patient care.
The AI-driven insights provided by the system facilitate proactive healthcare management, enabling
early intervention based on behavioural patterns or irregular health metrics. This data-centric

9 | 39
approach streamlines operations within healthcare facilities, minimizes human error, and improves
resource allocation.

Additionally, the platform promotes collaboration between families, healthcare professionals, and
organizations. By offering shared access to patient data and alerts, the platform fosters coordinated
care, ensuring all stakeholders are aligned in their efforts to safeguard the well-being of elderly
individuals. This interconnected approach strengthens trust among caregivers and enhances the
overall efficiency of elder care services.

From a social perspective, the platform empowers elderly users by promoting their autonomy while
ensuring their safety. The ability to define and customize geofenced zones allows older individuals to
move freely within safe areas, reducing feelings of confinement and enhancing their quality of life.
The platform also reassures family members, who can monitor their loved ones remotely and receive
real-time updates, alleviating anxiety and stress.

Furthermore, the platform addresses broader societal challenges by offering a scalable and cost-
effective solution for elder care. As the demand for caregiving resources increases, the platform
reduces dependence on human caregivers by supplementing their efforts with advanced technology.
This not only alleviates strain on healthcare systems but also makes high-quality care accessible to a
wider demographic, particularly in underserved communities.

The platform’s adherence to data protection laws such as POPIA ensures that user privacy is
respected, promoting ethical use of technology in healthcare. By prioritizing transparency in data
usage and system functionality, the platform builds trust among users, fostering acceptance and long-
term adoption.

On a developmental level, the platform’s implementation encourages technological literacy among

users, particularly the elderly, their families, and healthcare workers. Training sessions and hands-on
interactions with the system enhance digital skills and confidence in technology, contributing to
societal progress in embracing digital health solutions.

2 DESIGN GUIDELINES AND APPROACH

2.1 GENERAL OVERVIEW

10 | 39
The application operates as a comprehensive platform designed to enhance the safety and well-being
of elderly individuals, particularly those with cognitive impairments such as dementia or Alzheimer's
disease. This platform effectively integrates real-time monitoring and management functionalities,
balancing the independence of seniors with the critical need for safety.

For supply-side users, which include caregivers and healthcare providers, the platform offers several
key functionalities. Real-time tracking capabilities enable these users to monitor the exact location of
elderly individuals through GPS. This feature allows caregivers and healthcare providers to view
movement history and receive instant alerts if seniors enter or exit predefined safe zones. The
platform supports the configuration of customizable geofences, allowing users to define specific
areas and set up notifications for any breaches of these boundaries.

Additionally, the platform provides access to valuable behavioural data, including detailed movement
patterns and activity levels. This data is crucial for healthcare providers to track changes in behaviour
and identify potential health issues early. Caregivers can also manage alerts and notifications by
configuring thresholds for various scenarios, such as unauthorized exits or unusual activity patterns.
Integration with existing healthcare systems ensures that data can be shared seamlessly, facilitating
coordinated care. Furthermore, the platform offers reporting and analytics tools to support initiative-
taking health management and informed decision-making.

On the demand side, the platform caters to elderly individuals by offering a user-friendly interface
designed for ease of use, despite potential cognitive impairments. The application allows seniors to
share their location with caregivers and emergency contacts, ensuring safety while maintaining their
autonomy. It provides guidance for navigating within safe zones, including visual and auditory cues to
help users stay within defined boundaries. Privacy controls are also a significant feature, giving
elderly users the ability to manage who can access their location and behavioural data.

Emergency alerts can be manually triggered by the elderly users if they feel unsafe or need
assistance, notifying their caregivers and emergency contacts promptly. The platform is designed to
deliver a personalized experience, with settings and notifications adjusted based on individual
preferences and needs. This personalization ensures that each user’s unique requirements are met
while maintaining a high standard of safety and support.

In summary, this application serves as a versatile platform that effectively integrates functionalities
for both supply-side and demand-side users. By providing robust monitoring and management tools
for caregivers and healthcare providers, alongside user-friendly features for elderly individuals, the

11 | 39
platform creates a secure and supportive environment. It facilitates effective care management while
enabling seniors to maintain their independence, addressing the complex needs of modern elder
care.

2.2 GOALS, GUIDELINES, AND RISKS

2.2.1 SECURITY

Security is paramount in any application, particularly those dealing with sensitive information, such
as a prevention of wandering app for the elderly. The application must safeguard user data against
unauthorized access, breaches, and other forms of cyber threats. This involves implementing robust
encryption, secure authentication mechanisms, regular security audits, and ensuring data integrity.
By prioritizing security, the application will not only protect users' sensitive information but also build
trust and comply with legal requirements like POPIA.

2.2.2 PRIVACY

Privacy is a fundamental right that must be respected in the design of your application. Protecting
user privacy involves ensuring that personal data is collected, processed, and stored in compliance
with privacy laws and regulations. The application should limit data collection to what is strictly
necessary for its functionality, provide transparency about data usage, and offer users control over
their personal information. By embedding privacy into the design, the application aligns with the
principles of data minimization and user consent, ensuring that users' privacy rights are upheld.

2.2.3 LAWS OF IDENTITY PRIVACY

Designing an application that respects privacy, and security involves carefully considering various
principles, especially those outlined in Kim Cameron’s Laws of Identity Privacy (2010). Two critical
laws from Cameron’s framework that are particularly applicable to the design of a prevention of
wandering app for the elderly are User Control and Consent and Minimal Disclosure for a Constrained
Use. These principles are essential for ensuring that the application not only protects user data but
also aligns with legal standards like the Protection of Personal Information Act (POPIA).

User Control and Consent:

The first law, User Control and Consent, is foundational to building trust between the user and the
application. Cameron emphasizes that users should have ultimate control over their personal

12 | 39
information, deciding when and how it is shared. This law highlights the importance of informed
consent, ensuring that users are fully aware of what data is being collected, how it will be used, and
who will have access to it. In the context of the prevention of wandering app, this principle can be
implemented by designing the application to be transparent about its data collection practices from
the outset. Upon installation, users should be presented with clear, concise privacy policies that
explain what data the app will collect (such as location data), why it is necessary, and how it will
enhance their safety or the safety of their loved ones. This transparency ensures that users are
making informed decisions about their privacy.

Minimal Disclosure for a Constrained Use

The second law, Minimal Disclosure for a Constrained Use, is equally critical. This law posits that
identity systems should only disclose the minimal amount of information necessary for a particular
transaction or interaction. This principle is grounded in the concept of data minimization, which is
also a key requirement under many data protection regulations, including POPIA.

For the prevention of wandering app, this law would translate into a design that limits data collection
to only what is essential for the app’s functionality. For instance, while the app may need to track an
elderly person’s location to prevent wandering, it should do so in a way that minimizes unnecessary
data retention or sharing. The app should avoid collecting additional personal information, such as
health records or unrelated personal identifiers, unless it is necessary for the app’s purpose.
Additionally, the app should provide users with clear options for what data they are willing to
disclose. For example, users might choose to only share location data when they are outside of a
designated safe zone, rather than continuously. This not only reduces the amount of data collected
but also gives users a sense of control over their personal information, further enhancing trust in the
application.

2.2.4 POPIA, PAIA, AND OTHER APPLICABLE REGULATORY FRAMEWORKS

The Protection of Personal Information Act (POPIA) is a comprehensive South African law that
governs the collection, processing, storage, and dissemination of personal information. Its primary
goal is to protect individuals' privacy rights by regulating how their data is handled. In the context of a
wandering prevention app designed for the elderly, compliance with POPIA is crucial to ensure that
the application is both legally compliant and respectful of user privacy.

13 | 39
 0. Lawful Processing of Personal Information

POPIA mandates that personal information must be processed lawfully and transparently. This means
that the wandering prevention app must obtain explicit consent from users before collecting their
personal data. For example, the app may need to collect location data to monitor the movements of
elderly users, providing real-time updates to caregivers. Before this data is collected, the app must
clearly inform users (or their legal guardians) about what information will be gathered, how it will be
used, and who will have access to it.

To comply with this requirement, the app should include a clear and accessible privacy policy that
outlines its data processing activities. Users should be required to provide informed consent, ideally
through an opt-in mechanism, before any data collection begins. This not only fulfils the legal
obligation under POPIA but also aligns with Cameron’s User Control and Consent principle, ensuring
that users maintain control over their personal information.

b. Data Minimization and Purpose Limitation

The data minimization concept is another essential component of POPIA compliance. The wandering
avoidance software needs to make sure that the least amount of personal information is gathered
and used for it to function. This is consistent with Cameron's principle of Minimal Disclosure for a
Constrained Use, which highlights the idea that an identification system should only reveal the
minimal amount of identifying data needed for a given purpose.

For instance, unless it is strictly required for the app's operation, the app should refrain from
collecting extraneous, irrelevant personal information, such as financial information or social
interactions, even though it may need to track a user's position to prevent straying. The app should
also use robust data anonymization methods to make sure that any information shared with outside
parties—like emergency responders—is devoid of superfluous personal identification information.

Additionally, only clearly defined, reasonable, and precise objectives may be for which personal
information is gathered under the POPIA's concept of purpose limitation. This indicates that location
data should only be utilized to prevent wandering in the context of the wandering prevention app.
Without further user consent, the app cannot utilize the data it has obtained for secondary uses like
marketing or profiling.

c. Security Safeguards

POPIA mandates that appropriate security measures be put in place to protect personal data from
threats like loss, unauthorized access, destruction, or damage. Strong security measures must be put

14 | 39
in place by the wandering prevention software to safeguard the confidential information it gathers,
especially location data that might be abused if it ends up in the wrong hands.

End-to-end encryption of data transfers, safe server storage of confidential information, and frequent
security audits to find and fix flaws are a few examples of security measures. Secure authentication
techniques, such two-factor authentication, should also be included in the app to guarantee that only
authorized users can access it and the sensitive data it handles.

In addition to being required by law, integrating POPIA compliance into the wandering prevention
app's design and functionality is essential for fostering user confidence and guaranteeing the app's
long-term viability. The app will preserve user privacy and personal information while also setting a
high bar for privacy and security in the digital health space by abiding by the principles of lawful
processing, data minimization, purpose limitation, and security protections. By taking these steps, the
app will uphold its ethical and legal requirements while guaranteeing the privacy and dignity of the
senior users it seeks to safeguard.

2.2.5 TRUST & (OR VS) TRANSPARENCY

Trust and transparency are fundamental pillars for the success of the wandering prevention platform.
These aspects ensure that users, caregivers, and healthcare professionals feel confident in the system
and its functionality, fostering widespread adoption and engagement.

Trust is built through the platform’s consistent performance, reliability, and commitment to
safeguarding user data. Users need to trust that the system will operate as intended, providing
accurate geofencing alerts, real-time notifications, and reliable health monitoring. Furthermore,
caregivers and healthcare professionals must have confidence in the data integrity and the AI-driven
insights provided by the platform. This requires robust testing, quality assurance, and adherence to
industry standards.

Data privacy and security play a critical role in establishing trust. By complying with data protection
laws such as POPIA, GDPR, and HIPAA, the platform ensures that sensitive user information is
handled with the highest levels of confidentiality and protection. Implementing encryption, secure
data storage, and stringent access controls reinforces this trust, particularly given the sensitive nature
of health and location data.

15 | 39
Transparency, on the other hand, complements trust by providing clarity about how the platform
operates. Users need to understand the system’s functionalities, the purpose of data collection, and
how their data will be used. This includes clear communication about AI processes, such as how alerts
are generated or how behavioural patterns are analysed. Providing detailed terms of service and
user-friendly privacy policies ensures that users are fully informed about their participation.

Transparency also extends to how the platform handles errors or system limitations. Users should be
notified of any technical issues, such as potential downtime or limitations in geolocation accuracy, to
manage expectations and maintain confidence. For healthcare professionals, transparency in
reporting and analytics helps ensure that clinical decisions are based on accurate and interpretable
data.

The relationship between trust and transparency is interdependent; one cannot exist effectively
without the other. Trust is strengthened when users perceive the system as transparent, while
transparency alone is insufficient if the system lacks reliability. By balancing both, the wandering
prevention platform creates a secure, ethical, and user-friendly environment that addresses the
needs of all stakeholders and supports its adoption as a trusted tool for elder care.

2.2.6 AREAS OF RISK

2.2.6.1 LEVERAGE EXISTING ARCHITECTURE

One significant risk is the platform's reliance on integrating with existing infrastructure and
technology. Compatibility issues may arise when attempting to connect with legacy systems, such as
outdated caregiver tools or healthcare facility management software. If the existing architecture is
not robust or scalable, it might impede the performance of the platform. Furthermore, integrating
real-time tracking systems into older architectures may require costly upgrades or lead to system
instability, reducing the platform's efficiency. These challenges emphasize the importance of
conducting thorough compatibility testing and using modular designs to minimize disruption.

2.2.6.2 DEVELOPMENT ENVIRONMENT

The choice of development environment poses another risk. If the environment lacks stability,
scalability, or support for modern tools and frameworks, the platform's performance and

16 | 39
maintainability may be compromised. Moreover, using a poorly configured development
environment may result in inefficiencies, bugs, or delays in deployment. Development environments
also need to support collaborative workflows, particularly in distributed teams, to ensure seamless
project progress. Establishing a reliable, flexible, and well-documented development environment is
essential for mitigating these risks.

2.2.6.3 EASE OF USE

The platform's success depends heavily on its ease of use for both elderly users and caregivers. A
complex or unintuitive user interface can lead to poor adoption rates, decreased user satisfaction,
and higher training costs. Seniors, who may have limited familiarity with digital devices, require
simple navigation and clear instructions. Similarly, caregivers need tools that enhance their
workflows without adding unnecessary complexity. If the platform fails to prioritize usability, it risks
alienating its target audience. Usability testing and iterative design processes should be employed to
ensure accessibility and user-friendliness.

2.2.6.4 EXTENSIBILITY

A rigid platform architecture that does not accommodate future growth or integration with emerging
technologies poses a significant risk. As technology evolves, the platform may need to integrate with
new AI tools, wearables, or healthcare systems. If the platform lacks extensibility, these integrations
could become challenging, leading to obsolescence or limited functionality over time. A modular,
service-oriented architecture is crucial to ensure that the platform remains adaptable and future-
proof, mitigating risks associated with technological stagnation.

2.2.6.5 API ENABLED

A critical component of the platform's functionality is its ability to interact with other systems
through APIs. If the APIs are poorly designed, lack robust documentation, or fail to adhere to
standard protocols, integration challenges could arise. These issues may limit the platform’s
interoperability with third-party devices, healthcare management systems, or caregiver applications.
Security risks, such as unauthorized access or data breaches, may also emerge from insufficiently
secured APIs. Therefore, API design should follow best practices, including authentication,
encryption, and scalability considerations, to minimize these risks.

2.2.6.6 RESTFUL FRAMEWORK

17 | 39
The reliance on a RESTful framework, while advantageous for scalability and standardization,
introduces potential risks related to system performance and security. If the framework is not
properly implemented, it could lead to latency issues, especially when handling large volumes of real-
time data from tracking devices. Additionally, improper handling of HTTP methods, errors, or security
measures within the RESTful framework could expose the system to vulnerabilities. To mitigate these
risks, the development team should adopt industry-standard practices, including efficient routing,
rate limiting, and robust error handling within the RESTful framework.

3 REQUIREMENTS & DESIGN

3.1 GENERAL OVERIEW

The wandering prevention app is designed to enhance the safety and independence of elderly
individuals, particularly those with cognitive impairments like dementia or Alzheimer’s disease. The
app integrates real-time tracking and monitoring functionalities, enabling caregivers and healthcare
providers to maintain oversight while allowing seniors to move freely within designated safe zones.

Business Requirements: The app must provide robust features such as real-time location tracking,
geofencing capabilities, behavioural data monitoring, alert management, emergency alerts, and
privacy controls. It should facilitate seamless integration with existing healthcare systems to ensure
efficient data sharing among caregivers, healthcare providers, and family members.

Functional Requirements: For supply-side users (caregivers and healthcare providers), the app should
feature a dashboard displaying real-time location data, movement history, and alerts for when
seniors exit safe zones. Users must be able to configure customizable geofences and set thresholds
for alerts based on scenarios like unauthorized exits. For demand-side users (elderly individuals), the
app needs a user-friendly interface that allows them to trigger emergency alerts, navigate within safe
zones using visual and auditory cues, and manage privacy settings regarding data access.

Assumptions and Constraints: The design assumes varying levels of technological proficiency among
users, especially elderly individuals, thus prioritizing simplicity and ease of use. Constraints include
compliance with privacy laws like POPIA and implementing robust security measures to protect
sensitive data. Additionally, the app must function reliably in areas with limited internet connectivity,
requiring offline capabilities and local data storage.

18 | 39
 3.2 BUSINESS REQUIREMENTS

Requirement
Real-Time Location The app must provide accurate, real-time GPS tracking of elderly users,
Tracking allowing caregivers to monitor their whereabouts at all times.
Geofencing Users should be able to create and customize geofences (virtual
Capabilities boundaries) around safe zones, receiving alerts when seniors enter or exit
these zones.
Behavioral Data The app should collect and analyze movement patterns and activity levels to
Monitoring identify changes that may indicate cognitive decline or potential health
issues.
Alert and Caregivers must have the ability to configure alerts for specific scenarios,
Notification such as unauthorized exits or unusual activity, to ensure timely responses.
Management
Emergency Alert The app must allow elderly users to manually trigger emergency alerts,
Functionality notifying caregivers and emergency contacts if they feel unsafe or need
assistance.
Privacy Controls The app should offer robust privacy settings, enabling elderly users to
manage who can access their location and behavioral data.
Integration with The app must facilitate seamless data sharing with existing healthcare
Healthcare Systems systems to enhance coordinated care and ensure continuous monitoring of
users.

3.3 FUNCTIIONAL REQUIREMENTS (FOR BOTH SIDES OF PLATFORM)

Requirem Descripti Requirement No. Description

ent No. on

FR1 User Authentication: Secure login for elderly users or their family
members.

FR2 Geofencing: Ability to track user movements and alert when leaving safe

19 | 39
 zones.

FR3 Real-Time Alerts: Notifications to family members when geofence

boundaries are breached.

FR4 Health Monitoring: Integration with wearable devices for vital signs
tracking.

FR5 Behavioral Data Visualization: User-friendly display of activity patterns

and health trends.

FR6 Customizable Preferences: Allow users or family members to set

notification types and urgency levels.

FR7 Data Privacy: Ensure compliance with data protection laws (e.g., POPIA) to
safeguard personal data.

Requirem Descripti Requirement No. Description

ent No. on

FR1 User Authentication: Secure login for healthcare professionals with role-
based access.

FR2 Geofencing Management: Ability to set and modify geofenced areas for
patients.

FR3 Real-Time Alerts: Notifications for boundary breaches and abnormal

health patterns.

FR4 Health Data Integration: Access to real-time and historical health metrics
from wearable devices.

FR5 Behavioral Data Analysis: AI-driven insights into patient movement and
activity patterns.

FR6 Reporting and Analytics: Tools for generating and analyzing patient
reports.

20 | 39
 FR7 Device Management: Monitoring the status and connectivity of wearable
and tracking devices.

FR8 System Scalability: Ability to manage multiple patients and devices

simultaneously.

3.4 ASSUMPTIONS/CONSTRAINTS/DEPENDENCIES/RISKS

The development and implementation of the wandering prevention platform rest on several
assumptions, constraints, dependencies, and potential risks, each of which plays a critical role in
ensuring the system's effectiveness and sustainability.

A key assumption is that all users, including patients, caregivers, and healthcare professionals, will
have access to smartphones or compatible devices to interact with the platform. Reliable internet
connectivity is also assumed to be available, facilitating real-time data synchronization and alert
notifications. Additionally, wearable devices used for health monitoring are expected to have
sufficient battery life to support continuous operation without frequent interruptions. Furthermore,
it is presumed that caregivers and healthcare professionals will receive adequate training to use the
platform effectively, enabling them to maximize its features and provide better care.

The platform operates under certain constraints, including strict adherence to data protection laws
such as POPIA, GDPR, or HIPAA. These legal requirements necessitate robust privacy safeguards,
potentially influencing the system's design and data handling processes. Resource limitations during
the initial deployment may restrict scalability, confining the platform to a smaller user base until
further funding or resources become available. The accuracy of geolocation, a critical feature of the
platform, is constrained by GPS technology, which can face limitations in indoor or remote
environments. Additionally, the integration of third-party APIs, such as Google Maps for geofencing
and navigation, is subject to the terms, policies, and availability of those services.

Several dependencies must also be considered for the platform to function optimally. The system's
compatibility with medical equipment, such as heart rate and temperature monitors, is essential for
health monitoring features. Secure cloud storage is another critical dependency, ensuring that user
data is managed and analysed without compromise. The platform also relies on continuous support
and updates for wearable device firmware and mobile applications to maintain functionality and
address emerging challenges. Furthermore, the use of external services, including Google Maps, for

21 | 39
geolocation and navigation functionalities, is a significant dependency that must be maintained to
provide accurate and reliable service.

The platform's success also depends on managing several risks effectively. One major risk is the
potential for data breaches, which could lead to unauthorized access to sensitive user health and
location data, undermining user trust and compliance with legal requirements. System downtime,
caused by technical failures or service disruptions, poses another significant risk, as it can lead to gaps
in monitoring and compromise user safety. Resistance to adopting new technology, particularly
among elderly users who may struggle with navigating modern systems, is a potential barrier to
widespread acceptance and effectiveness. The malfunctioning of wearable devices or tracking
equipment could disrupt operations and affect the reliability of the platform. Lastly, there is a risk of
over-reliance on technology by caregivers, which could result in diminished personal attention to
patients and reduced quality of care.

By identifying these assumptions, constraints, dependencies, and risks, the design and
implementation of the wandering prevention platform can be tailored to address these challenges
proactively, ensuring a reliable and user-centered system that meets the needs of all stakeholders.

3.5 USER SCENARIOS

1. Producer Using System (Caregiver/Healthcare Provider)

Mr. Thompson, an elderly patient with early-stage dementia, is looked after by Maria. She keeps an
eye on his whereabouts using the senior wandering prevention app. She examines the geofence she
put up around his house when she accesses the app's dashboard on Monday morning. Maria uses the
real-time location tool throughout the day, and it indicates that Mr. Thompson is securely in his
allotted area. Maria gets a notification from the app at midday letting her know that Mr. Thompson
has left the secure area. She looks up his position on the map right away and sees that he's two
blocks away from the park. Maria makes sure Mr. Thompson gets home safely by calling him on the
app. She logs a note in the app about the incident, helping her keep track of his wandering patterns
for future care adjustments.

2. Consumer Using System (Elderly User)

22 | 39
To assist secure his safety, Mr. Thompson, an older man suffering from early-stage dementia, has a
wandering prevention software put on his phone. He receives a gentle notice from the app that he is
getting close to his safe zone border one morning while out on a stroll. Mr. Thompson, not knowing
where he is, hits the "Help" button on the app, which tells his caregiver Maria where he is. Soon after,
Maria gets in touch with him, and he uses the app's navigation hints to assist him follow the
instructions back to his safe area. With gratitude, Mr. Thompson moves on with his day, at ease in
the knowledge that assistance is only a tap away thanks to the app.

3. Analyst/Reporter Using System (Healthcare Provider/Administrator)

Dr. Ahmed is a healthcare analyst who tracks patterns among senior patients with cognitive decline
using the wandering prevention app. He enters the system and looks over the information that has
been collected from several patients—including Mr. Thompson—during the previous month. A report
outlining behavioural changes, such as the number and timing of boundary violations, has been
created by the app. According to the statistics, Mr. Thompson has been more roving throughout the
early afternoon. Dr. Ahmed advises adding additional afternoon activities and modifying Mr.
Thompson's medication regimen in order to lessen restlessness. He communicates this knowledge
with Mr. Thompson's caretakers. Dr. Ahmed may improve patient care by using the app's reporting
features, which are based on actual behavioural patterns.

4 ARCHITECTURE DESIGN

4.1 SYSTEM ARCHITECTURE

4.1.1 WEB APPLICATION

The purpose of the wandering prevention app is to serve as a central monitoring point for healthcare
professionals and caregivers of older people who are prone to straying. With its user-friendly
interface (UI) and live GPS map, caregivers may monitor activities and set up geofences—virtual
boundaries that sound an alarm when they are crossed. In order to provide individualized care,
caregivers can update medical records, manage profiles, and alter alert sensitivity settings.
Through RESTful APIs, the back end facilitates communication between the user interface and
databases, processes requests, and manages data in real-time. The back-end records event, provides
notifications, and guarantees data consistency when a geofence is activated or location data is

23 | 39
requested. Encryption and authentication procedures protect sensitive data and guarantee POPIA
compliance, making data security a top priority.

By stressing user control and limited disclosure, users can maintain privacy through minimal data
gathering and control over what information is disclosed. This aligns with Kim Cameron's Laws of
Identity Privacy. Additionally, the software interfaces with current healthcare systems, facilitating
data sharing for coordinated treatment and early cognitive decline diagnosis.

This architecture puts functionality, privacy, and safety first. It offers a seamless, secure user
experience and helps caregivers take care of the elderly.

4.1.2 MOBILE APPLICATION

The mobile app system architecture of the wandering avoidance software is designed to provide real-
time tracking, smooth data synchronization, and offline functionality. The Figma-created user
interface places a high value on simplicity, ensuring that big, clear buttons and simple navigation are
accessible to older adults. Flutter and React Native are examples of cross-platform frameworks that
offer consistent functionality between iOS and Android devices.

Caregiver Functionality:

When elderly people leave geofenced zones, caregivers receive instant notifications. This is a benefit
of real-time location tracking. They may modify notifications, change geofence settings, and manage
user profiles. Push alerts guarantee that caretakers are made aware of situations as soon as possible,
allowing for timely response.

Elderly Users Functionality:

The software lets senior users set off emergency alarms that instantly inform caretakers, as well as
passive geolocation tracking. To help their everyday routines, health reminders are also incorporated,
such as medication prompts.

Front-End: Simplified Cross-Platform Interface

For iOS and Android compatibility, the mobile app will be developed utilizing frameworks like React
Native or Flutter. Its Figma-designed user interface (UI) prioritizes simplicity and ease of use, with
features like big buttons, clear notifications, and easy navigation to make it suitable for older users
and caretakers. All users are guaranteed accessibility and seamless engagement as a result.

Back-End: Cloud APIs for Real-Time Sync

24 | 39
The application handles data processing, synchronization, and real-time updates, such as location
data and alarms, using cloud-based APIs. Modifications, including altered geofence configurations,
are immediately synchronized with every device. In order to keep caregivers informed, push
notifications provide instant warnings for geofence breaches or crises.

Local Storage for Offline Use

When offline, the app caches important information like location and warnings using local storage. It
immediately synchronizes with the server upon reconnecting, guaranteeing that caretakers can keep
an eye on things and get updates even in places with poor connectivity.
All things considered, the architecture guarantees a dependable, expandable, and easy-to-use
solution for older users as well as caretakers.

4.2 HARDWARE ARCHITECTURE

The wandering avoidance software will leverage cloud-based hosting using Microsoft Azure for
scalability, security, and accessibility. Cloud infrastructure lowers upfront server costs with a pay-as-
you-go model and allows for dynamic scalability, guaranteeing consistent performance even as user
demand increases. Cloud infrastructure is much more affordable as a result.

Caretakers and medical professionals may access real-time data from anywhere, which makes
remote monitoring and coordinated care possible. The platform will secure sensitive data by
enforcing data encryption and conducting security audits to ensure compliance with HIPAA and
POPIA regulations. Users will have control over how they access data thanks to user management
mechanisms.

Furthermore, cloud hosting facilitates seamless interaction with healthcare systems, opening the
door to efficient data exchange for preventative healthcare. Furthermore, backup systems and
disaster recovery methods provide data security and recovery in the case of a catastrophe.

4.2.1 CONNECTIVITY

The majority of the app's features, such as geofencing alerts and real-time location monitoring,
depend on internet access. To guarantee usability when offline, through:

25 | 39
 Offline Functionality

Critical data, like safe zone borders and last known position, will be locally stored by the app. Once
the device is connected again, alerts will be queued and sent out.

4.3 SOFTWARE ARCHITECTURE

Algorithms that can learn from data and generate inferences or predictions without explicit
programming are referred to as machine learning algorithms. Machine Learning will be used in the
following function:

Anomaly Detection

Finding odd or strange patterns in an older person's behavior or movement that might indicate
wandering is known as anomaly detection. This might be accomplished in the app by employing
machine learning to identify instances in which a person's movement deviates from their typical,
predictable patterns, perhaps signaling danger.

Specific Machine Learning Algorithms:

a) Isolation Forest: This technique separates data points at random and determines whether or
not they are isolated from the remainder of the dataset in order to find abnormalities. The
likelihood of a point being an anomaly (i.e., wandering behaviour) increases with the number
of partitions needed to isolate it.
b) One-Class Support Vector Machine (SVM): This algorithm identifies deviations as possible
anomalies by learning the traits of typical behaviour. It's especially helpful when there are lots
of data on regular behaviour but few instances of wandering incidents.
c) Autoencoders (Neural Networks): Autoencoders compress data and make an effort to recreate
it; they are used for unsupervised anomaly detection. When it comes to wandering avoidance,
a significant reconstruction mistake may indicate an abnormality or unanticipated movement.

Using one of these algorithms, the app would continually analyse data from many sources, such
as motion sensors and GPS, to identify anomalies in real-time.

2. Predictive Health Care

26 | 39
By examining trends in health data, predictive healthcare forecasts possible problems or hazards.
Based on past data and behavioural tendencies, this element of the wandering avoidance software
might be used to identify circumstances in which the senior user may be more likely to stray or fall.

Specific Machine Learning Algorithms:

a) Random Forest: To forecast future events, this ensemble learning technique constructs many
decision trees and combines them. It may be used to forecast health hazards like
disorientation or cognitive deterioration, which could raise the risk of straying. The program
could interpret data from several inputs, including historical wandering events, heart rate,
and motion patterns.
Logistic Regression: Based on behavioural and health data, logistic regression can determine if a
person is at danger of straying within a given time period. It is useful for binary classification
problems.

c) Recurrent Neural Networks (RNNs) and Long Short-Term Memory (LSTM): These algorithms are
perfect for time-series data and may anticipate future health risks or wandering episodes by
modelling sequences of health-related behaviours (e.g., sleep patterns, medication
adherence). When it comes to capturing long-term dependencies in behavioural data, LSTMs
work very well.

3. Geofence Optimization

The process of geofence optimization is improving the virtual barriers, or geofences, that set off
alarms when they are crossed. By getting to know a person's movement habits and making
adjustments accordingly, optimized geofences increase accuracy and prevent missed or false
warnings.

Specific Machine Learning Algorithms:

a) a) K-Means Clustering: By analysing past location data, this unsupervised learning

method may be utilized to find common regions where the elderly person usually
moves. The clusters might be useful in determining the best geofence bounds.
b) b) Reinforcement Learning: Using information from past breaches and safe routes, this
technique might dynamically modify geofences. To reduce false positives (unnecessary
warnings) and false negatives (missed wandering occurrences), the system would

27 | 39
 optimize the placement of geofences. Through feedback—whether alarms are
activated or not—the algorithm becomes increasingly adept at adjusting geofences.
c) A different clustering method that can determine a probability distribution for common
sites is the Gaussian Mixture Model (GMM). This approach is useful for developing
more adaptable geofence borders that take into account both the likelihood that a
person would be in a certain region and the actual distance.
d) The program can make sure geofences respond to the user's typical behaviour patterns
and offer the best protection against wandering by using machine learning.

4. Specific Machine Learning Algorithms

These algorithms provide the basis of the fundamental features previously mentioned:

a) Decision Trees: Based on a sequence of inquiries regarding the user's activity or health
data, these straightforward, comprehensible algorithms may be utilized to make
judgments. When interpretability is crucial for caretakers, they are very helpful.

b) Support Vector Machines (SVMs): This method works well for classification jobs and
can distinguish between typical and anomalous behaviour (anomalies) depending on a
range of characteristics such movement speed, location, and time of day.

c) Deep Neural Networks (DNNs): DNNs are a type of deep learning model that can
uncover intricate patterns and correlations from huge and complicated datasets (such
continuous GPS tracking or multi-sensor data) that standard models would overlook.
They are very helpful for anomaly detection and predictive healthcare.

d) Bayesian Networks: Based on partial data, these probabilistic models may be used to
make predictions in uncertain situations, such determining the chance of wandering.
They can also simulate causal links, providing information about the conditions that
may encourage wandering.

The wandering avoidance software uses machine learning algorithms to continually learn from real-
world data and improve its performance while intelligently monitoring movement, anticipating
hazards, and optimizing its functioning for a smooth user experience.

28 | 39
 4.4 DATA & INTERNAL COMMUNICATION

4.4.1 DATA & INTERNAL COMMUNICATION

What data will be generated by your application, and by which user groups?

a) Elderly people's GPS whereabouts and mobility patterns are included in the location data,
which is created in real-time as they move around. Geofencing and anomaly detection depend
on this data. The target audience is experienced users.
b) Geofence Information: The boundaries or personal safety zones that each senior's carers have
drawn. The target user group is caregivers.
c) Alerts and Warnings: Real-time alerts are sent out in the event that a geofence border is
crossed or if unusual movement is seen. This includes the location, time, and kind of alert of
the occurrence. The intended user group includes both older users and caregivers (should
they receive notifications on their devices).
d) Health and Behaviour Data: By analysing behavioural patterns, such as activity levels, sleep
patterns, or shifts in mobility that suggest cognitive decline or a higher risk of straying,
predictive healthcare algorithms will produce insights. Senior users are the intended user
group.
e) Usage Data: Interaction logs that show how users—elderly people and caregivers—interact
with the program, such as when they alter settings, accept notifications, or use it often.
Elderly users and caregivers are the intended user group.

4.4.2 WHERE WILL THIS DATA BE STORED?

a) Cloud Storage: The app’s data (location, geofence, alerts, health patterns, etc.) will be stored
in a secure cloud-based database (e.g., AWS, Google Cloud, or Microsoft Azure). This ensures
real-time synchronization across devices and accessibility from multiple platforms.
b) Local Storage: To allow offline functioning, a subset of the data, such as recent movement
patterns or geofence settings, will be cached locally on the user's device. Restoring connection
will cause this data to sync with the cloud once more.
c) Encrypted Databases: To protect user privacy and security, sensitive data (such as behavioral
and health data) is encrypted both in transit and at rest.

4.4.3 WHAT DATA MANAGEMENT PROCESSES SHOULD BE PUT IN PLACE TO

MANAGE THIS DATA?

29 | 39
Several data management processes should be implemented, including:

a) Data Encryption: All sensitive user data (location, health, personal information) should be
encrypted during storage and transmission to prevent unauthorized access.
b) Access Control: Implement role-based access control (RBAC) to ensure that only authorized
users (e.g., caregivers, healthcare professionals) can access specific data. User authentication
and multi-factor authentication (MFA) should be used.
c) Data Retention Policies: Establish policies to define how long data will be stored and when it
will be deleted or archived. This is essential to comply with privacy regulations (e.g., GDPR,
HIPAA) and to ensure that old or irrelevant data is properly disposed of.
d) Data Synchronization: Ensure that data synchronization between local devices and cloud
storage is seamless and happens in real time when connectivity is available. Mechanisms for
conflict resolution (e.g., when the same data is edited offline and online) should be
implemented.
e) Backup and Recovery: Regular backups of cloud-stored data must be performed to protect
against data loss. An efficient recovery plan should be in place to restore data quickly in case
of accidental deletion or system failures.
f) Audit Trails: Maintain logs that track changes to the data (e.g., geofence modifications, alert
acknowledgments), providing transparency and the ability to review historical data.

4.4.4 WHAT ADDITIONAL VALUE WOULD PLATFORM OWNERS BE ABLE TO

EXTRACT FROM THIS DATA?

The data generated by the app can provide several insights and opportunities for platform owners:

a) Behavioural Insights: By analysing movement patterns and health data, platform owners can
identify trends in elderly individuals' behaviour, which could inform the development of new
features, such as fall detection or improved health monitoring.
b) Predictive Analytics: Aggregating data across multiple users could lead to more accurate
predictive models for wandering behaviour, allowing platform owners to offer advanced
healthcare insights or partner with healthcare providers.
c) Geolocation Analytics: Platform owners could aggregate geofence and location data to
optimize geofencing algorithms, providing better wandering prevention accuracy and
reducing false alarms.

30 | 39
 d) User Engagement Metrics: Analysis of app usage data could reveal how caregivers and elderly
users interact with the app, helping to refine the user experience and improve customer
satisfaction.
e) Monetization Opportunities: By anonymizing and aggregating the health and location data,
platform owners could create reports or insights that could be shared with healthcare
organizations, insurers, or research institutions, provided users consent to the data sharing.
f) Personalization: The data can also be used to personalize the app experience, offering
caregivers tailored notifications, or recommendations based on the user’s health and
behaviour data.

5 SOLUTION DESIGN

5.1 VISUAL DESIGN

5.1.1 WIREFRAMES & VISUALS

5.1.1.1 MOBILE APPLICATION

This section presents six meticulously designed User Interface (UI) screens that comprise our
wandering avoidance prototype's general design approach. Fluid UI was used in the development of
these screens. A thorough explanation of each screen's role is included, emphasizing the features
that it will make possible for the The launch screen, which aims to make supply and demand sides of the
a good first impression, is the first point
of contact that users have with the
market. wandering prevention software. This
screen, which presents the brand
identity through a simple and eye-
Launch screen catching design, acts as the app's entry
point.

It has the app's logo placed The launch screen serves as the app's
prominently in the centre, over a introduction and is essential in
background that is neutral purple establishing the tone for the entire
shades with a pop of colour user experience by providing a
(turquoise) and conveys smooth transition into the main
dependability. The deliberate features of the application.
simplicity of the design makes it easy
for consumers to identify the app and
gives them confidence in its intended
usage.

31 | 39
Log-In screen

32 | 39
 Home screen The wandering prevention app's
home screen is made to be simple to
use and provide instant access to
important functions. Essentially, an
easily navigable map gives users a
clear picture of their actual location,
enabling family members and
caregivers to keep an eye on them in
This dashboard is an essential tool for real time.
efficient monitoring and
administration since it allows you to
view previous data, set up and
manage alarms, and modify
wandering prevention system
settings. The user-friendly design
makes sure that all necessary Top searchers are prominently
features are easily accessible, highlighted, such as an emergency
improving users' safety and locator for quick response in an
convenience. emergency and prescription
notifications for timely reminders to
take prescribed drugs. Furthermore, a
thorough dashboard overview
provides a single point where supply-
side users, such caregivers or
healthcare providers, can engage
with different app features.

Screen 4: Emergency contact

The screen for sending alerts and adding
emergency contacts is made to offer a simple and
effective method of handling important
notifications. By providing names and phone
numbers, users may quickly add emergency
contacts, ensuring that the appropriate individuals
are notified in the event of an emergency. The
interface includes an easy-to-use form for entering
and safely storing contact information.

There is a large "Send Alert" button on the screen in

case an older person gets lost or begins to wander.
When this button is touched, the current location
and any pertinent information about the situation
are sent, along with an instant message to any
designated emergency contacts. The alert system's
dynamic and user-friendly design ensures that
assistance may be promptly summoned when
needed, giving users peace of mind.

33 | 39
 Screen 5: Emergency Locator

The screen intended for supply side personnel,

such caregivers or medical professionals, is an
essential tool for handling cases of wandering.
An alarm is immediately issued to the selected
contacts when an older person is seen to be lost
or roaming. This alert gives users immediate
notice of the problem by being displayed
prominently on the screen. The user interface has a "Find" button
that, when pressed, uses the built-in
map to determine the elderly
person's exact location in real time.
Because of the map's simple
navigation, users can find and track
the person quickly. This feature
makes sure that medical
professionals and caregivers can
quickly and effectively coordinate
their reaction, improving safety and
addressing possible straying
problems.

Screen 6: User Management

Clear sections and tabs that convey

facts in an approachable way make
the interface easy to navigate. By
offering a consolidated view of all
pertinent patient data, this feature
The user management screen is an promotes efficient monitoring and
extensive interface created to management and makes rapid
simplify patient profile maintenance interventions and individualized care
and oversight. This page gives supply possible.
users—like caretakers or medical
professionals—access to
comprehensive profiles of every
senior in their care. Viewers can
select a profile and then get access to
a variety of vital data, such as
prescription specifics to guarantee
treatment compliance, wandering
statistics that monitor trends and
events, and real-time position
updates via a map.

34 | 39
 5.2 DATABASE DESIGN

5.2.1 ENTITY RELATIONSHIP DIAGRAM

5.3 DATA CONVERSION AND DATA ANALYTICS

Data conversion and analytics are integral to the functionality of the roaming prevention platform.
The platform must seamlessly ingest, process, and analyse data from diverse sources to ensure
effective monitoring and decision-making. Data conversion involves transforming raw data collected
from tracking devices, geofences, and medical equipment into standardized formats suitable for
analysis. This process includes data cleaning, formatting, and integration to maintain accuracy and
consistency.

Data analytics, on the other hand, plays a pivotal role in extracting actionable insights from the
collected information. The platform employs advanced AI and machine learning techniques to
analyse behavioural patterns, predict potential wandering incidents, and generate alerts. Predictive
analytics enhances proactive caregiving, while trend analysis supports long-term healthcare planning.
The integration of data analytics also allows for the generation of reports on caregiver response
times, system efficiency, and user health statistics. Additionally, anonymized data can be utilized for
research purposes, fostering advancements in elder care and dementia management.

35 | 39
 6 EXTERNAL INTERFACE DESIGN

6.1 GOOGLE MAPS

The platform integrates with Google Maps for real-time geolocation tracking and geofencing
capabilities. Google Maps provides accurate mapping data to define safe zones for elderly users and
track their movements. The interface allows caregivers to set geofence boundaries easily and receive
alerts when these boundaries are breached. Its robust API ensures reliable navigation and location-
based services, enhancing the platform’s functionality.

6.2 MEDICAL EQUIPMENT

Integration with medical equipment is a key feature of the platform, enabling comprehensive health
monitoring alongside location tracking.

6.2.1 HEART RATE & BLOOD PRESSURE MONITOR

The platform connects with wearable devices that track heart rate and blood pressure. These devices
provide real-time health data, which the platform analyzes to detect anomalies. Sudden spikes or
drops in these metrics can trigger alerts, prompting caregivers to intervene promptly.

6.2.2 TEMPERATURE MONITOR

Wearable or external temperature monitoring devices integrate with the platform to track body
temperature continuously. This feature ensures that caregivers are alerted to potential health issues,
such as fever or hypothermia, which are common concerns in elderly care.

6.3 GENERAL OFFICE EQUIPMENT

The platform also supports integration with general office equipment to facilitate communication,
documentation, and reporting.

6.3.1 PRINTER/SCANNER/EMAIL/NOTIFCATIONS

The platform enables caregivers and administrators to print or scan reports and documents related to
the elderly user’s health and location data. Additionally, it supports email notifications and alerts,
ensuring caregivers and family members are informed promptly. Push notifications on mobile devices

36 | 39
further enhance responsiveness, ensuring that critical information reaches the right individuals
without delay.

By integrating these external systems and interfaces, the platform ensures comprehensive
monitoring, efficient communication, and a user-friendly experience, aligning with its goal of
providing safe and reliable elder care solutions.

7 LIVING LABS DEMONSTRATION AND FEEDBACK

7.1 TEST GROUPS

TG No. Name Description

TG1 Healthcare Doctors, nurses, and caregivers responsible for monitoring elderly users'
Professional health and safety.
TG2 Customers/Patients Elderly individuals with dementia or cognitive challenges who benefit from
the roaming prevention platform.

7.1.1 TEST GROUP 1: HEALTHCARE PROFESSIONALS

Healthcare professionals, including doctors, nurses, and caregivers, form the first test group for the
living labs demonstration. This group focuses on assessing the platform's usability, functionality, and
integration within a healthcare setting.

Purpose of Testing:
To evaluate how effectively the platform supports healthcare professionals in monitoring and
managing elderly patients. Key aspects include ease of use, responsiveness of the alert system, and
integration with medical equipment.

Demonstration Approach:
The platform will be demonstrated in a controlled environment where professionals can simulate
real-life scenarios, such as setting geofences, responding to alerts, and analyzing health data.

Feedback Focus Areas:

 Interface Design: Clarity and accessibility of dashboards and tools.

 Alert System: Timeliness, accuracy, and relevance of alerts.

37 | 39
  Medical Integration: Functionality of health monitoring devices, such as heart rate
and temperature monitors.

 Workflow Impact: How well the platform integrates into existing care processes
and reduces workload.

Expected Outcomes:

 Identification of potential design improvements or technical gaps.

 Validation of the platform’s value in enhancing patient safety and caregiving

efficiency.

7.1.2 TEST GROUP 2: CUSTOMER/PATIENT

The second test group includes elderly individuals who will use the platform, as well as their family
members and informal caregivers.

Purpose of Testing:
To evaluate the platform’s user-friendliness and its impact on the safety and well-being of elderly
users.

Demonstration Approach:
Elderly users will be equipped with the platform’s tracking and monitoring devices in a simulated
home environment. Family members or caregivers will interact with the system to observe its
functionality, including alerts and geofencing features.

Feedback Focus Areas:

 Ease of Use: Accessibility of features for elderly users, considering limited technical
proficiency.

 Geofence Alerts: Reliability and clarity of notifications sent to caregivers.

 Safety and Comfort: Perception of security and autonomy among elderly users.

 Family Satisfaction: Peace of mind provided to family members regarding the

safety of their loved ones.

Expected Outcomes:

 Insights into the platform’s effectiveness in real-world scenarios.

38 | 39
 Feedback on user experience to refine design and functionality.

 Increased understanding of user needs to enhance the platform’s adaptability.

39 | 39