Case Studies AI Demand Planning Reduces Inventory Carrying Costs, Improves Inventory Turns, and Credentials Supply Chain

AI Demand Planning Reduces Inventory Carrying Costs, Improves Inventory Turns, and Credentials Supply Chain

AI Supply Chain Optimization

The furniture retailer was struggling with excess carrying costs and missed revenue due to inaccurate forecasts and poor supply chain visibility. To address these challenges, we developed an AI-driven demand planning system that accurately predicts demand for thousands of products over multiple weeks. This automated system has improved forecasting accuracy by 15%, resulting in annual savings of $16 million across 120 stores. By implementing this innovative solution, the supply chain team has regained trust within the business leading to improved strategic decision-making and enhanced operational efficiency.

Critical Issue

The mid-sized furniture retailer faced significant challenges, including $34 million in excess carrying costs and missed revenue due to stockouts. The supply chain analysts’ inconsistent and inaccurate forecasts, coupled with poor visibility and governance, led to bad predictions of consumer demand, resulting in erratic inventory levels and unaccounted over/under-buy costs. The perpetuation of these patterns of mistakes undermined business operations, eroded trust in the supply chain, and hindered the responsible forecasting process.

Operation Icon

Customer Profile

Mid-sized furniture retailer

120 stores in 16 states

Crisis Icon

Key Problems

$34M in excess carrying costs

Missed revenue from stockouts

our solution

We developed a demand planning system that accurately predicts 10-to-26-week demand for 10,000 SKU/DCs, addressing long-standing business issues such as intermittent, erratic, lumpy, and smooth demand patterns. This fully automated system is currently running in production and includes a user-friendly dashboard for monitoring accuracy, as well as intuitive explanations for the forecasts. The supply chain team was trained on how to effectively use the system, and it was implemented in just 10 weeks by a team of one. The technology stack used for this system includes Azure, Python, Azure Data Factory, ADLS, AutoML, and Databricks.

The results

Schedule Icon

Our system has successfully saved the company

$16 million annually across 120 stores by being

15% more accurate than human forecasts, leading to improved estimates of consumer demand. This significant cost savings not only enhances financial performance but also accelerates the effectiveness of new supply chain analysts in the industry, allowing them to be proficient in their roles more quickly.

lightbulb with a gear inside of it

Our solution is facilitating the accumulation of

institutional knowledge by enabling the learning,

encoding, and sharing of the drivers of the business

processes. As a result, the Supply Chain team is

rebuilding trust with the business, leveraging this

knowledge to improve strategic decision-making

and enhance overall operational efficiency.

Explore Our AI Solutions

Interested in exploring how our AI solutions are creating net-new revenue streams?

Concurrency Center of Excellence

Learn more about Concurrency Centers of Excellence online