Restaurant Chain Streamlines IT Resolution Across Thousands of Locations Using Generative AI
About the Customer
The customer is a leading American fast-food chain with over 2,700 restaurants, known for its commitment to high-quality ingredients and exceptional service. Headquartered in the Southeast, it stands out for its strong corporate culture, franchise model emphasizing owner-operator involvement, and community-focused initiatives. With steady growth and a loyal customer base, it remains a significant player in the fast-food industry.
The Customer's Challenge
As the restaurant chain expanded, so did the volume of IT inquiries from its growing network of locations. Without AI solutions, the business faced two costly options: either expand the IT department to manage the increasing inquiries or allow response times to lag, potentially disrupting operations and impacting customer experience.
With an average of 185 new locations added each year, the company needed a more scalable solution to handle IT inquiries efficiently. The previous approach was labor-intensive and time-consuming for both the IT department and franchise locations across the country.
phData's Solution
As part of a leadership-driven, company-wide initiative to leverage AI in all parts of their business, the restaurant chain needed a conversational natural-language interface to service the IT needs across its network of 2,700 restaurants. phData was contracted to execute a solution and proposed building a custom help-bot webpage powered by a vector database of the customer’s IT documentation on AWS.Â
With the goal of reducing IT phone calls and the back-and-forth of clarifying emails, the Retrieval Augmented Generation (RAG) solution would aim to pull relative IT documentation quickly and efficiently when prompted by a restaurant location.Â
The three components that needed implementation by phData to create the solution were:
- An IT documentation ingestion pipeline
- A vector database
- Help Bot Webpage
IT documentation from ServiceNow is vector encoded and stored in a vector database. Interfacing with the vectorized IT documentation is executed during the invocation of an AWS Lambda function, as well as the Mistral Large text generation model hosted on Amazon Bedrock.Â
Amazon API Gateway serves as the conduit between the LLM-based chat interface and the vector database, all architected by phData within a secure AWS account. RAG was used to prevent hallucinations and tailor chat responses based on enterprise-specific knowledge from IT support documentation.
An architectural diagram of phData’s solution
Results
After exploring five different LLMs in an interchangeable interface for comparison and evaluation, phData settled on Mistral Large on Amazon Bedrock as the most capable for the customer’s IT requirements. phData was able to successfully vector-encode more than 10,000 (and growing) IT documents for retrieval, setting the foundation for an enterprise-specific solution. As a result of the newly introduced AI application by phData, the resolution time for common IT questions was 30x faster than the customer’s previous IT workflows. Moreover, time previously spent on common IT questions could be spent on more pressing and demanding IT issues across locations.Â
Even with the continued expectation of additional locations across the US and beyond, the generative AI solution will allow the customer to scale comfortably and effectively without growing its technical costs to accommodate growing IT needs.Â
The solution was a cost-effective way for the company to start utilizing AI within its organization and with phData showing the incremental value of AI, the organization has since added additional bots and plans to accelerate future AI projects.Â
LLMs deployed behind an interchangable interface
IT service documents vector-encoded for retrieval
Faster resolution for common IT questions
Why phData?
phData is the leading AI and data services company. We specialize in AI and data applications, from conception to production. Our global delivery team partners with the world’s top brands to execute data initiatives in artificial intelligence, data engineering, applications, analytics, and managed services for cloud platforms.Â
Take the next step
with phData.
Learn how phData can help solve your most challenging data analytics and machine learning problems.
