ITC Infotech acquires Blazeclan Technologies to enhance Multi-Cloud services and fast-track digital transformation

GPTW Badges - PNG - NOV 2022_NOV 2023

A Leading Pizza Restaurant Chain Boosts its End-Customer Experience through Predictive Time Model for Accurate Order Estimates

About Customer

The customer is a leading pizza restaurant chain in Malaysia and focuses on being the best pizza delivery firm. With over 240 pizza places located strategically across the region, the customer is increasing its stores for serving all pizza lovers in the nation. It continues to grow as a major ecommerce entity and has led several firsts in its digital platforms.

The Challenge

The customer wanted to achieve absolute accuracy for their processes, right from pizza preparation to delivery. They wanted to develop a predictive time system for calculating the pizza order ready time, wait time, and arrival time.

Accuracy of the prediction time was supposed to equal the actual process timings, considering the delays caused due to several factors such as spike in orders, issues with pizza preparation and process, unpredictable delivery-bound circumstances, etc. Depending on the predictive time model, the customer sought to improve the processes, thereby enhancing customer experience and the overall business.

Key requirements included

  1. Calculating actual time of delivery to the consumers​
  2. Notifying the consumers about delays
  3. Notifying the delivery people about delays in picking up the orders based on predicted pizza preparation time

The Solution

The customer partnered with Blazeclan to develop the predictive time model for improving their consumer experience. Understanding the customer’s processes, rules of business logic, and the way they calculate time for order ready, wait, and arrival was a major challenge. Another key challenge was testing the model in the dummy environment by adhering to all the aforesaid scenarios.

Blazeclan developed the predictive time model to meet the customer’s requirements effectively using a rule-based framework. Major factors were considered for achieving high consumer experience, such as information from the customer, historical data, and near-real-time calculated data.The model was built keeping in mind all three stages of the order life cycle, namely,

1. Order Ready Time – This included scenarios of the unhidden as well as hidden orders.
2. Order Wait Time – This included scenarios where the order is ready without the delivery person to pick it up or the order is ready with the delivery person available right on time.
3. Order Arrival Time – This included scenarios where the delivery person arrives at the customer’s door before/right on time or the rider returns to the store.

Simplified Diagram of the Deployed Predictive Time Model

Benefits Achieved by the Customer

  • The call center staff has access to the pizza arrival prediction time, which helps them in responding to consumer queries about their order’s status.​
  • Store managers, using the pizza arrival prediction time, can accurately determine the estimated time of an order’s arrival to the consumer.
  • The delivery person, with access to the prediction time, can ensure timely delivery of the orders.
  • Having the foresight of predicted pizza delivery life cycle helped the customer in achieving high efficiency across their business operations.

Tech Stack

AWS Lambda Amazon API Gateway Amazon CloudWatch
Amazon EventBridge​ AWS Parameter Store​ MongoDB
Visual Studio Code​ Git/GitLab​ Postman​
SonarQube​ Node.js