About the client
RS Components, the number one high-service distributor of electronics and maintenance products across Europe and Asia Pacific.
Challenge
As a part of a strategic goal to be the market leader in customer experience, RS Components needed an IoT platform that would engage customers and drive sales, right down to the factory floor at customer sites. The platform needed to be pushing the boundaries of digital innovation and offer services like Image Recognition and AI-based natural language product search.
Key goals
- Embed tacit knowledge in accessible technology.
- Enable frictionless sales on production sites.
- Create edge-computing capability within customers plants.
- Drive engagement with the wider RS digital ecosystem
Solution
Hack & Craft led the process, from Ideation to Rollout and Continuous Delivery, using a combined hardware and software agile approach focused on the needs of RS Components.
The ConnectPoint software, built specifically for this project, includes a unified UX/UI that provides a seamless experience for Image Recognition and AI product search. Our design & product teams also created a bespoke plinth design that houses the IoT components and lightbox needed for Image Recognition.
How we did it
We formed an innovation team with key stakeholders from RS Components and experts from Hack & Craft. This innovation team took part through all stages of the iterative development process, playing a key role in steering the solution for best market fit.
Our project team leveraged their extensive expertise and experience in providing machine learning and Image Recognition, including a bespoke Machine Learning toolkit and automated test harnesses.
Approach
Hack & Craft used a unique four phase agile methodology (Discovery, Development, Roll Out and Continuous Delivery & Support), allowing incremental refinements based on stakeholder and real-world feedback.
The design & product team conducted Discovery sessions to map out user stories and inform plinth designs, before first creating 2D and then 3D concepts for feedback.
Our hardware team then produced the 3D Engineering CAD files, bill of materials and design for the electronic components and wiring schemes.
In parallel software teams implemented the user stories and ConnectPoint app on the custom IoT hardware with a touchscreen based UX/UI, Image Recognition and AI search.
Tech
Hack & Craft delivered a multi-layered software solution, from hardware programming for device drivers enabling 4G cellular connectivity to the machine learning and AI models.
Users are able to search the RS catalogue by uploading an image through a multi-layered Image Recognition framework that utilises TensorFlow and ResNet-50 convolutional neural networks. Hack & Craft’s AI team further augmented this with bespoke colour analyser and OCR machine learning layers.
The text product search is powered by AI natural language processing models, which utilise self-learning through continued end user use.
Integration of communication channels include real-time live chat with Customer Support and a bespoke SMTP email solution that provides messenger like communication with RS Local Branches.
The plinths support remote diagnostics and debugging with Over The Air update capability and in-depth analytics providing long-term insights such as demand indicators and sales-conversion.
Impact
The plinths are a clear market differentiator that directly place the full RS catalogue and communication channels on to the factory floors of customer sites. Driving both revenue growth and increased customer retention by facilitating the identification, search and purchase of products from a central location.
The distributed IoT platform that has been built also creates a foundation for additional disruptive Industry 4.0 tools and services that can be rapidly and cost efficiently rolled out.