How long does it take to deliver an order a customer has placed? Is the in-progress order going to meet the timelines or is it at risk of missing timelines? These are some of the common questions any organization has when dealing with customer orders.
Business Challenge
One of our Telecom clients had to ensure that they delivered all orders within the delivery timeline. The order delivery process involved several departments – Network, Service Delivery, and Finance. They faced challenges in delivering fixed broadband solutions which involved just configuration to maybe even building a new fiber route.
Our challenge was to collect data from different departments and large variations in timelines and sometimes a network expansion was required before even the order was delivered. These posed major challenges while predicting delivery timeline estimation.
Approach and Solution
- Our approach was to use the ML-based Order Delivery Timeline Estimation Model developed on Databricks that estimates when a consumer will receive their order.
- The entire order is divided into several tasks, and the model is trained using historical data to predict how long each task will take to complete.
- Historical Data from various sources are ingested in Azure Synapse.
- As the order progresses, the time taken to complete each task is compared against the initial prediction to understand delays and risks in order delivery timelines. The total time required to finish the order is then calculated by adding up all the tasks.
- This enables managers to take early corrective action and keep orders on schedule thus reducing delays.
- Thus this model predicts when a new order will be delivered as soon as it is placed.
- A react-based dashboard was provided to visualize the real-time status of Orders, delays, and progress.
- The entire solution is built and deployed in Azure.
Benefits
- End-to-end view of various tasks in order and identification of bottlenecks in the process.
- Proactive action on delayed tasks to ensure overall order completion within timelines improving customer satisfaction.
Technology used
Azure, Azure Synapse, Azure Databricks, ReactJS
Network
Capacity Planing
Our client had thousands of servers spread across the globe carrying internet traffic. The main challenge was managing these servers at optimal utilization.
Vendor Contract Analytics
Our client had thousands of contract documents scanned and stored as pdf documents. Manually scouring through these documents...
CDN
Log Analytics
Our client receives a large number of crowd sourced images. They receive many pictures which are blurry and some of the images...