AI as a helpful assistant in a workplace audit
The Story
For Safety managers, preparing for an audit or inspection is usually stressful and time-consuming. They have to go searching for documents, verify incident logs, and piece together what happened if things weren't documented perfectly from the beginning.
From use case to reality
By leveraging generative AI, we wanted to build a product that could instantly parse through their dense documentation and logged data, pulling up the exact information and specific details needed to be audit-ready on demand.
Key Outcomes
Decrease the financial cost of audits by helping employees to be audit-ready with minimal external support
Minimise nonconformities reported by anticipating and addressing potential issues prior to the audit
Shorten the length of audits by reducing the retrieval time of information requested mid-audit
AI Model
Using HybridRAG technology for unstructured, multimodal data
We needed to know exact dates, equipment, employee data, safety procedures, etc and the relationships between them. Because the source material came in different formats, the proposed model used HybridRAG to combine semantic search, keyword matching, OCR, and relationship mapping.
The system could find the right evidence, understand how it connected, and return information that users could review before using it in an audit response.
Connection model showing how related safety information can be linked to give AI more context.
Workflow Mapping
From request to review-ready response
Once the workflow is initiated, the AI would help gather the right records and connect the available evidence. The safety managers could then review, ask follow-up questions, and build a clearer picture of what happened before preparing the final response to auditors.
Connection model showing how related safety information can be linked to give AI more context.
User Flow Explortion
Turning the AI model into a usable review flow
User flows helped us visualise the exact steps a manager takes and helped translate the AI logic into practical screens, states, and decisions.
Early user flows to help connect the AI model to the users steps
Step 1
The safety manager enters the information they are trying to gather, and the system begins finding the records and evidence that could support it.
Step 2
The intake form captures the core case details so the system has enough context to retrieve relevant details and build a case.
Step 3
Once the case details are submitted, the system begins parsing available records and collecting the documents connected to the incident.
Step 4
The AI parses through all the available records and presents the case details, collected documents, and an initial summary in one place, with all the sources available in the side panel.
Step 5
Any area can be investigated further by asking a more specific question within the same workspace, with follow-up questions building on what has already been found, rather than sending the user back to a new search.
Step 6
Once a follow-up question is submitted, the AI re-parses the relevant documents and provides a more focused summary, with the available evidence presented upfront.
