Sharper Supply Chains With Smart AI Modeling

Our client, one of the largest global players in the manufacturing and industrial sector, faced a significant challenge in streamlining their inventory management process for their largest customer, an e-commerce giant.
The client needed to overcome significant challenges in various SKU inventory levels fueled by an erratic demand while maintaining a 99% fill rate, leading to substantial excess inventory costs and customer service issues.

Cutting Down Inventory Carrying Costs
Our client had been grappling with the issue of excess inventory costs. The conventional approach they’d been following at their warehouses involved maintaining an inventory level of between 14 and 21 days. This approach led to higher holding costs and lower fill rate, as well as operational inefficiencies resulting from overstocked warehouses.

Dicing Demand Variability
A significant challenge for our client was the erratic and unpredictable nature of demand across various SKUs from a prominent customer. Traditional forecasting methods, which often relied on historical data and established patterns, led to a heuristic approach to inventory stocking. As a result, our client struggled to cope with this level of demand variability that was marked by sharp fluctuations, seasonal variations, and unforeseen shifts in product demand preferences.

Chopping Up Supply Chain Complexity
Confronted by intricate distribution channels and a diverse product portfolio, our client realized that streamlining the supply chain was a top priority across all their SKUs. Without any data analytics, they struggled with tackling logistical hurdles, coordinating warehouse manpower, optimizing production, and managing deliveries. Their primary target was to achieve cost-effectiveness and responsiveness, ensuring they could meet the evolving demands of their customer effectively.

Our solution followed a structured approach and comprised of three phases:

The first phase involved defining data requirements, exploring data sets, and conducting initial inventory and forecast analysis.

Secondly, the team consolidated data from various sources, extensively analyzed forecast accuracy, demand patterns, highest revenue SKUs, and inventory costs, identified opportunities for improvement, and did an error analysis. In the final phase, the team constructed inventory models, automated data pipelines, and meticulously documented the entire process before handing it over to our client’s internal analytics team.

These phases collectively drove the successful implementation of a data-driven inventory optimization solution.

What we provided: A finely chopped inventory analysis model
The Min-Max Inventory analysis model developed for our client used a data-driven approach to determine optimal inventory levels. The main inputs into the model were the historical shipment data, the previous 13 weeks of forecast data, the expected demand for the next 13 weeks, the expected service level defined at 99%, and the inventory on hand at warehouses.

By incorporating these parameters, the model calculates the minimum and maximum inventory thresholds for each SKU-Plant combination. This gave our client’s demand planners much-needed insight to maintain efficient stock levels, thereby reducing excess inventory carrying costs.

The model is automated, facilitating real-time adjustments and enabling responsive inventory management to navi gate demand fluctuations effectively.
Fractal’s in-depth knowledge in supply chain helped decode the specific challenges in the existing process and allowed for a more targeted and effective solution.

-Priyasmita Ghose Engagement Manager, Consulting Industrials, Energy and Travel

The immediate impact
By implementing the Min-Max Inventory analysis model, our client realized a remarkable 46% reduction in inventory levels. Simultaneously, the 99% service level was consistently maintained, enhancing customer satisfaction. The successful transformation of inventory management allowed the demand planners of our client’s largest customer to not only streamline warehouse operations but also significantly improve manpower utilization.
Long-term benefits
Fractal’s analysis model can be scaled and implemented seamlessly across our client’s customers – from small local retailers to massive global distributors. It acts as a blueprint, helping our client consistently align inventory levels with fluctuating demand, as well as optimize manpower utilization and streamline warehousing operations.

The client provided Fractal an NPS score of 10 and said that A slice of clarity through inventory transformation the biggest impact of their partnership with Fractal was “improved supply chain efficiency”