Case Study

GenAI Assistant for Asset Monitoring

Designing a generative AI Chatbot that helped speed up enterprise platform configuration by 200%
Overview

What is Intelligent Assets?

Intelligent Assets is an IoT asset-monitoring app for industrial, large-scale use cases.

Why does IA need an intergrated AI Assistant?

A major challenge with IA is ensuring user adoption and a smooth up-front setup process. Since customers have tons of assets to track, they don’t want to spend tons of time adding each one to IA. We created a GenAI Assistant built on an LLM to help make the process of creating new asset types, assets, event types, and other essential components much faster for customers who were monitoring asset at large scales.

Solution Overview

(For 1st Release MVP)

  1. User enters a prompt --> AI Assistant completes the requested task.

  2. AI Assistant can generate the following items upon request:

    1. Asset types

    2. Assets

    3. Event types

    4. Rules (pre-cursors to event types and events)

  3. AI Assistant can also answer general questions about Intelligent Assets.

What is Intelligent Assets?

Intelligent Assets is an IoT asset-monitoring app for industrial, large-scale use cases.

Why does IA need an intergrated AI Assistant?

A major challenge with IA is ensuring user adoption and a smooth up-front setup process. Since customers have tons of assets to track, they don’t want to spend tons of time adding each one to IA. We created a GenAI Assistant built on an LLM to help make the process of creating new asset types, assets, event types, and other essential components much faster for customers who were monitoring asset at large scales.

Solution Overview

(For 1st Release MVP)

  1. User enters a prompt --> AI Assistant completes the requested task.

  2. AI Assistant can generate the following items upon request:

    1. Asset types

    2. Assets

    3. Event types

    4. Rules (pre-cursors to event types and events)

  3. AI Assistant can also answer general questions about Intelligent Assets.

THe PROBLEM

Before the AI Assistant, setting up a new Intelligent Assets system was a pain in the @#!$.

Customers wanting to start connecting their assets to the cloud would have to manually create each asset type, event type, rule, etc. via IA's standard "Create X" settings menus. This process proved to be daunting and time-consuming for most of our target audiences. Oftentimes, they'd give up before they even started–or give up halfway through after getting confused about what certain settings meant–and request help from one of our internal Services Engineers, whose primary job was to assist customers with all their setup and system oversight tasks.

Before the AI Assistant, setting up a new Intelligent Assets system was a pain in the @#!$.

Customers wanting to start connecting their assets to the cloud would have to manually create each asset type, event type, rule, etc. via IA's standard "Create X" settings menus. This process proved to be daunting and time-consuming for most of our target audiences. Oftentimes, they'd give up before they even started–or give up halfway through after getting confused about what certain settings meant–and request help from one of our internal Services Engineers, whose primary job was to assist customers with all their setup and system oversight tasks.

TIMELINE

2 months

TEAM

  • 1 Junior UX/UI Designer

  • 1 AI Full-Stack Developer

  • Product & Services Stakeholders

TOOLS

Figma, Figam, Figma AI, Jira, Confluence, Slack, Claude, ChatGPT, Heurio

MY ROLE

Lead UX/UI Designer

Solution Preview

Phase 1

Phase 1

Requirements & Scope

  • Project Kickoff meeting(s) - typically w/ services engineers, ux-team engineers, and the designer (me [and now you])

  • Understand business requirements

  • Define problem statement

  • Understand the users

  • Stakeholder interviews (with Services and/or other team-members)

  • Reference prior interview notes/insights

Turning Interviews Into Insights

Audit existing solution

Users avoid using certain filters due to complexity or insufficient context.
Users are requesting multi-selection & “select all” in categorical fields

Heuristic Analysis

Check in with devs?

Competitor Analysis

Personas

Phase 2

Define

User Personas

Competitor Analysis

Information Architecture

Breaking Things Down:
Filter Data Types

User Flows for Each Data Type

Phase 3

Ideate

Our Ideation Process

Design Iterations

Phase 4

Test & Refine

Test & Refine

Preparing the User Test

User Testing Results:

Post-testing Revisions

Phase 5

Deliver

Reflections & Next Steps

Overview

What is Intelligent Assets?

Intelligent Assets is an IoT asset-monitoring app for industrial, large-scale use cases.

Why does IA need an intergrated AI Assistant?

A major challenge with IA is ensuring user adoption and a smooth up-front setup process. Since customers have tons of assets to track, they don’t want to spend tons of time adding each one to IA. We created a GenAI Assistant built on an LLM to help make the process of creating new asset types, assets, event types, and other essential components much faster for customers who were monitoring asset at large scales.

Solution Overview

(For 1st Release MVP)

  1. User enters a prompt --> AI Assistant completes the requested task.

  2. AI Assistant can generate the following items upon request:

    1. Asset types

    2. Assets

    3. Event types

    4. Rules (pre-cursors to event types and events)

  3. AI Assistant can also answer general questions about Intelligent Assets.

What is Intelligent Assets?

Intelligent Assets is an IoT asset-monitoring app for industrial, large-scale use cases.

Why does IA need an intergrated AI Assistant?

A major challenge with IA is ensuring user adoption and a smooth up-front setup process. Since customers have tons of assets to track, they don’t want to spend tons of time adding each one to IA. We created a GenAI Assistant built on an LLM to help make the process of creating new asset types, assets, event types, and other essential components much faster for customers who were monitoring asset at large scales.

Solution Overview

(For 1st Release MVP)

  1. User enters a prompt --> AI Assistant completes the requested task.

  2. AI Assistant can generate the following items upon request:

    1. Asset types

    2. Assets

    3. Event types

    4. Rules (pre-cursors to event types and events)

  3. AI Assistant can also answer general questions about Intelligent Assets.

Insight 1

Users avoid using certain filters due to complexity or insufficient context.

Insight 2

Users are requesting multi-selection & “select all” in categorical fields

Insight 3

Users want to use the search bar as an immediate way to filter

Insight 4

Main goal: Identify assets with abnormal data before things escalate.

Full Case Study Coming Soon!

Back to Home