Page 10 - Preliminary Analysis Towards a Standardized Readiness Framework - Interim Report

Preliminary Analysis Towards a Standardized Readiness Framework

the generalizability of AI/ML solutions across regions and domains via transfer learning.
Hardware implementations, especially open-source IoT boards are evolving to host the edge
data processing. Reference network implementations provided via OneM2M [22] reference
is maturing to the level of wide-scale deployments. IoT gateways such as LoRa gateway and
applications, SDKs, and APIs.
6) Data collection and model validation via Sandbox pilot experimental setups

Many use cases using data from the experimental sandbox, created experimental solutions and
validated them using experimental setups. While real-world data would imply a more reliable
source of data and a realistic testing environment, not all scenarios could be encountered in
the real world, mostly catastrophic events and related data are rare.

7) Storage and computing via Core Cloud and Edge Cloud

There are physical infrastructure elements such as speed bumps and barricades for traffic
regulations, in-vehicle safety accessories such as advanced driving assistance system (ADAS),
wireless sensor deployments, energy sources such as solar panels, and charging stations for
drones, which are important for reaping the benefits of AI-based solutions.

However, there is backend infrastructure, such as compute availability and storage availability
fiber/wireless availability for the last mile, and high-speed network capabilities, which would
democratize AI/ML solutions and create scalability for innovations.

This report captures 5 case studies in clause 3, which bring focus to specific aspects or impacts
of the readiness factors. The mapping of readiness factors is represented in figures which call
out the specific readiness factors which applies to that case study.

The case studies involve multiple use cases. This report covers 16 use cases from various
domains. Each use case may in turn have different use case scenarios. Clause 4 has a summary
of use cases whereas the detailed scenarios are called out in Appendix A and B.

Future work

Over the next months, the number of case studies, use cases, and scenarios would be scaled
to include a diverse set of domains and regions. We propose the future steps as the initial
building blocks towards a multi-dimensional, practical, and evolving framework, the “ITU AI for
Good Infinity framework for AI readiness”. A pictorial representation of the Infinity framework
is given below.

2