HVAC Servicing Company Reduces Technician Assignment Time by 50% with AI

Challenge

A California-based HVAC servicing company was facing a critical challenge as they grappled with the increasing volume of service requests, leading to slow and inefficient technician assignments. Manual review processes conducted by dispatch operators, involving considerations such as technician experience, availability, and
proximity to customer locations, proved to be time-consuming, error-prone, and resulted in prolonged service waiting times.

Our Approach

We developed an AI-powered recommendation engine that automates the process of assigning technicians to service requests.The recommendation engine uses a variety of factors to determine the best-suited technician for each request, including:

Technician’s experience with similar requests

Technician’s previous customers reviews

Technician’s proximity to the customer’s location

Technician’s availability

The recommendation engine was thoughtfully developed as an internal web app, ensuring seamless adoption by the company’s dispatch operators.

Deployment

The recommendation engine was developed using the following technologies:

SpaCy: A natural language processing library that was used to extract keywords from service requests

GPS data: Integrated with the company’s GPS IoT platform to find the driving distance of available technicians from the customer’s address

Learning-to-Rank algorithm: Trained on historical data to determine matching scores for available technicians given a service request ID

Flask: A Python microframework used to develop the backend of the recommendation engine

React: A JavaScript library used to develop the frontend of the recommendation engine

The entire application was deployed in Amazon Web Services (AWS) as a web app. The historical data used to train the Learning-to-Rank algorithm was stored in a Microsoft SQL Server database.

Business Impact

The implementation of the recommendation engine has resulted in a number of benefits for the HVAC servicing company, including:

REDUCED TIME TO ASSIGN TECHNICIANS:

The average time needed to assign a technician to a service request has been reduced by nearly half. This has freed up the company’s dispatch operators to focus on other tasks, such as customer service.

INCREASED CUSTOMER SATISFACTION:

The company’s customer satisfaction (NPS) score has increased due to the quicker allotment of technicians, as well as the superior matchmaking in terms of skill set, past experience, and current location of the technician.

Overall, the implementation of the Automated Technician Recommender Platform has been a success for the HVAC servicing company. The recommendation engine has helped the company to improve its efficiency, customer satisfaction, and bottom line.