Rajeev Gangwar
Case Studies

Selected Work

Each project tells a story of turning supply chain complexity into measurable business impact.

Applied Materials

End-to-End Supply Chain Simulation

Built discrete-event simulation models to replicate the global spare parts supply chain for semiconductor capital equipment.

The Challenge

Applied Materials operates a complex global spare parts network serving semiconductor fabs with near-zero tolerance for downtime. Traditional spreadsheet models couldn't capture the dynamic interactions between stochastic lead times, multi-echelon inventory policies, and capacity constraints across regional depots.

The Approach

Built an end-to-end supply chain simulation environment to model a global spare parts network from supplier to customer site. The simulation captures stochastic lead times, multi-echelon inventory policies, and capacity constraints across regional depots. Enables what-if scenario planning for disruption response, safety stock calibration, and capacity planning during demand spikes.

The Outcome

The simulation platform is now used for critical strategic decisions — from network realignment triggered by industry growth to disruption response planning. It revealed dynamics that static models completely missed, particularly around lead time variability and capacity constraints during demand spikes.

The Lesson

“Simulation beats spreadsheets every time when your supply chain has variability. A 15% demand increase doesn't cause a 15% inventory increase — the non-linear effects only become visible in simulation.”

DESPythonScenario Planning
Read full case study →
Logistics Network

$5M Tariff Mitigation Strategy

Redesigned the global repair network to mitigate 25% penalty tariffs on China-origin parts.

The Challenge

When 25% penalty tariffs were imposed on China-origin parts, the existing repair network routing sent parts through China-based facilities, exposing the business to millions in unexpected duty costs. The network needed rapid redesign without disrupting service levels.

The Approach

Developed import duty models to estimate total duty exposure across the network. Analyzed alternative routing through non-China facilities, balancing duty savings against logistics costs and cycle time impacts. Proposed network modifications that maintained service commitments while avoiding tariff-exposed routes.

The Outcome

$5M+ in duty savings achieved. The project was recognized with an AGS quarterly team award in 2019. The duty modeling framework continues to be used for ongoing tariff risk assessment.

The Lesson

“Trade policy changes create optimization opportunities, not just risks. The companies that respond fastest to tariff shifts turn compliance costs into competitive advantages.”

Network DesignDuty ModelingCost Optimization
Read full case study →
Warehouse Operations

Cartonization Algorithm for Outbound Packaging

Developed a box-size recommendation algorithm for airfreight packaging to reduce dimensional weight charges.

The Challenge

Outbound warehouse operations were using oversized boxes for airfreight shipments, resulting in excessive dimensional weight charges from carriers. Manual box selection by warehouse staff was inconsistent and suboptimal.

The Approach

Developed a cartonization algorithm that recommends optimal box sizes based on item dimensions, weight, and carrier-specific dimensional weight pricing rules. The algorithm considers multiple items per shipment and minimizes total shipping cost including both actual and dimensional weight.

The Outcome

$2.4M in savings from reduced dimensional weight charges. The algorithm was integrated into warehouse management workflows for consistent, automated box selection.

The Lesson

“Sometimes the biggest savings come from the most overlooked operational details. Nobody thinks about box sizes until you show them $2.4M in waste.”

Algorithm DesignPythonWarehouse Optimization
Read full case study →
Applied Materials

ML-Based Spare Parts Demand Forecasting

Developing machine learning models to forecast intermittent demand for semiconductor equipment spare parts.

The Challenge

Spare parts for semiconductor equipment exhibit classic intermittent demand patterns — long periods of zero demand punctuated by sporadic, high-value orders. Traditional forecasting methods like moving averages fail badly with this demand profile, leading to either chronic stockouts or excessive inventory.

The Approach

Comparing multiple approaches: Croston's method (the traditional intermittent demand baseline), gradient-boosted trees using part characteristics and installed base data, and ensemble methods that combine statistical and ML predictions. Building warehouse capacity forecasts using S&OP data and long-term growth trends.

The Outcome

Work in progress — early results show 30%+ improvement in forecast accuracy over traditional methods for key part categories. The forecasting models integrate with inventory optimization to automatically recommend safety stock levels.

The Lesson

“For intermittent demand, the installed base is your best predictor. Knowing which machines are in the field, their age, and their utilization patterns tells you more than years of demand history.”

Machine LearningPythonForecastingInventory
Read full case study →
RG
Ask Rajeev's AI
Online now