The Digital Transformation Challenge:
In the age of digital evolution, business leaders grapple with three pivotal questions:
Why is it imperative for companies to undergo digital transformation?
Where to initiate this transformation?
How to ensure the effective implementation of the transformation?
While many businesses jumped on the "Internet+" bandwagon, Taiwan's Wanglai Gas pioneered an O2O gas business with online orders and offline deliveries. But even with such innovative strategies, they couldn't outperform traditional gas companies. Was it due to their lack of industry experience? Rapid changes in internet dynamics? Or some other reasons?
Regardless, amidst the challenges posed by the pandemic economy, Wanglai sought to leverage digital transformation to break through its growth stagnation.
The Power of Data in Digital Transformation:
Digital transformation isn't a mere trend—it's the pathway to the future. It's about harnessing and maximizing the value of data, which could reshape traditional businesses into entirely new entities. Recognizing this potential, Wanglai pinpointed data as their weapon of distinction against competitors.
They already had a trove of data from their app and online orders—delivery routes, customer orders, member details, etc. Yet, this data had been manually collected, presenting challenges such as inaccuracy, limited types, low volume, lack of dimensionality, and isolation.
Digital Transformation Steps with JarviX:
Setting Business Goals:
The value chain is lengthy for most businesses—from R&D and supply chain management to marketing and logistics. To effectively digitize, companies must pinpoint the most impactful area of the business value chain. For Wanglai, it wasn't the manufacturing of gas, but rather, the highly variable and unpredictable delivery process. Their goal was to maximize delivery efficiency.
Wanglai's first step with JarviX was the creation of a control room. They aggregated data scattered across different systems—like delivery numbers, sales amounts, customer service reports, and online orders—into a single dashboard, often referred to as a data cockpit. With JarviX's platform, they could also obtain insights by simply asking questions like "What's the monthly forecast for inventory costs?" or "How much revenue did we generate this week?"
New data points like idle time, travel time, and distance traveled were incorporated to further refine their delivery process. This data was then formatted for machine learning models and underwent feature engineering.
Modeling & Solution Finding:
With a grasp on their delivery data, JarviX's AI began optimizing and creating a delivery simulator. This simulator could quickly develop a viable model and run millions of simulations to determine optimal vehicle numbers, routes, and delivery counts—direct factors impacting KPIs. Additionally, JarviX incorporated gas inventory monitoring. Whenever a customer ordered gas, the simulator would analyze nearby households for potential gas needs, optimizing delivery routes for proactive delivery.
The outcome? After optimizing for a period, Wanglai astonishingly reduced their delivery cost from $80 to just $30 per cylinder.