PepsiCo Uses FICO Optimization to Orchestrate Daily Production, Align Staffing and Accelerate New ideas

PepsiCo, founded in 1898, is one of the world’s largest food and beverage companies, with iconic brands like Fritos, Doritos, Gatorade, and Quaker sold in over 200 countries and territories. Guided by its “Winning with pep+” strategy, PepsiCo pairs global scale with a commitment to innovation, sustainability, and delivering positive impact worldwide.

PepsiCo has relied on FICO® Xpress Optimization since 2009 to tackle some of its most complex planning challenges. For more than 15 years, Xpress's mathematical optimization technology has become a core part of PepsiCo’s operations. With FICO Xpress, PepsiCo has:

• Cut daily production model solve times by 30% using Xpress’s solver tuning controls 
• Achieved an instant 15% reduction in solve time over the previous release by upgrading to the newest Xpress version 
• Deployed production scheduling globally across all manufacturing sites worldwide, including 36 sites across the United States and Canada 
• Solved 400–500 production planning instances per day across multiple planning horizons
• Reduced staffing plan creation from 2–3 days to less than 5 minutes for 80% of instances and less than an hour for all instances

Why Pepsico Uses FICO Xpress Optimization

• High performance at scale. PepsiCo runs extremely large multi-stage mixed-integer programming models with up to 400,000 constraints to plan production and staffing across dozens of sites, solving hundreds of scenarios each day. FICO Xpress Optimization delivers the speed and stability needed to support this volume and complexity, with most runs completing in under five minutes even on problems with millions of decision variables. This rapid performance allows schedulers to run multiple iterations each day and finalize plans with confidence. 
• Deep customizability. Unlike off-the-shelf enterprise scheduling systems that operate as black boxes, Xpress provides PepsiCo with the flexibility to build and continuously refine custom optimization models that adapt to its specific production realities and constraints. This ability to embed site-level logic has proven so effective that when PepsiCo evaluated switching to an enterprise ERP scheduling module, it found that system could only achieve only 40% of the accuracy of the Xpress-based model.
• Continuous improvement. PepsiCo upgrades to the newest Xpress release at least once a year to take advantage of new features and performance gains. These upgrades deliver immediate performance improvements allowing the company to compound performance gains over time and keep its daily production planning system aligned with evolving operational needs.
• Expert support. FICO’s technical team works closely with PepsiCo’s modelers to provide guidance on solver tuning, new feature adoption, and performance troubleshooting. When PepsiCo encountered solve time bottlenecks, FICO engineers helped identify and fine-tune solver parameters, significantly improving performance and strengthening confidence in the platform’s long-term viability.

This combination of performance, flexibility, continuous innovation, and trusted expert support has made Xpress PepsiCo’s go-to platform for solving mission-critical operational problems year after year.

Daily Production Planning: Orchestrating Production across a Global Network

PepsiCo must schedule production across a vast global network of manufacturing sites every day with speed and precision. Scheduling teams need to finalize plans by early afternoon and quickly rerun them whenever demand shifts or machines become unavailable.

This planning runs multiple times per day at 36 sites across the US and Canada, with similar deployments across Europe and Latin America, and is now used at all PepsiCo manufacturing sites worldwide. Across the network, planners execute 400-500 runs every day to keep operations on track.

Using FICO Xpress Optimization and Xpress Mosel, PepsiCo built a highly customized daily production planning application powered by a large-scale mixed-integer programming (MIP) model, containing up to 600,000 decision variables and up to 400,000 constraints. The company’s IT team developed the user interface that users engage with, which triggers the optimization model behind the scenes.

“I use Xpress Workbench to quickly debug modeling issues and develop enhancements for the master production scheduling model. Using FICO Xpress makes it easy to keep the model dynamic, so I can always add a new set of constraints without having to refactor the entire model. This allows us to cater the model to each sites’ specific needs. I appreciate the speed at which I can analyze an IIS and identify corrections needed to produce a feasible solution, which helps keep our planning capability functional.”

-- Brendan Schutte, Senior Scientist - Optimization at PepsiCo

The impact: faster planning, higher accuracy, global consistency

• Highly accurate schedules: The daily production planning model produces precise plans across dozens of sites and hundreds of runs per day, giving planners a dependable picture of what to produce and when.
• Faster decision cycles: Schedulers can now complete planning cycles earlier in the day and respond more quickly to unexpected changes in demand or capacity.
• Global consistency: The solution has been deployed worldwide, standardizing production planning practices across PepsiCo’s manufacturing network.
• Superior to off-the-shelf software: PepsiCo found that the daily production planning model delivered significantly better accuracy than enterprise scheduling platforms, validating the value of a fully customizable optimization solution.
• Proven performance gains: Upgrading to the latest Xpress release delivered a 15% reduction in solve time over the prior release, and solver tuning cut solve times by an additional 30%, allowing planners to run more scenarios and finalize plans faster.
• Proven longevity: After more than 15 years of continuous use, daily production planning remains a mission-critical application and a model for how PepsiCo approaches complex operational planning.

Headcount Optimization: Aligning Staffing with Operational Demand

PepsiCo must keep staffing levels aligned with fast-changing production demand across its manufacturing and warehouse sites. Hiring and onboarding new employees takes weeks, while frontline attrition can create sudden gaps in coverage.

PepsiCo built a predictive staffing tool using FICO Xpress Optimization with Xpress Mosel and Xpress Insight. The model uses mixed-integer and nonlinear programming to calculate how many full-time, part-time, and contract employees are needed and when to start hiring them based on expected demand and historical attrition patterns. It optimizes staffing across an entire site rather than individually by each department. 

The impact: faster planning and more resilient staffing

• Standardized practices: The solution replaced fragmented, site-specific spreadsheet planning with a consistent process deployed across multiple manufacturing and warehouse sites.
• Significant time savings: The tool reduced staffing plan creation from two to three days per department to less than an hour, freeing planners to reallocate their time to other important initiatives.
• Improved workforce planning: The model gives staffing teams immediate visibility into projected gaps, allowing them to start hiring earlier and prevent shortfalls.
• More efficient staffing mix: Optimizing the balance of full-time, part-time, and contract staff helps PepsiCo meet demand more flexibly while controlling labor costs.
• Better employee retention: By enabling more balanced and predictable scheduling, the tool has helped improve retention, especially among frontline workers.
• High user adoption: While initial skepticism slowed adoption, staffing teams embraced the tool once they saw its recommendations consistently outperform manual planning.

Proof of Concept Development: Accelerating New Optimization Ideas

“With Xpress, we can go from an idea to a working proof of concept in days, not months. That speed is what gets people excited about using optimization.”

--Lirthik Durai, Principal Scientist - Optimization at PepsiCo 

As more business teams across PepsiCo see the results of optimization, requests for new applications using Xpress continue to grow. But proving the value of an optimization use case can be time-consuming, and teams are often reluctant to adopt new innovations until they’re convinced the application is both accurate and reliable in the field. 

PepsiCo uses FICO Xpress Optimization and Xpress Mosel as a flexible development environment for PoC. The modeling team can stand up prototypes in days by reusing existing components and solver logic and then iterate rapidly with input from business stakeholders. Because Xpress solves large and complex models efficiently, teams can explore multiple approaches and scenarios early in the process without being blocked by performance limits. 

The impact: Faster innovation with lower risk

• Accelerated development: The modeling team can build PoCs in days instead of months, dramatically shortening the path from idea to solution.
• Lower upfront investment: Teams can more quickly validate the business value of their optimization projects before committing to full-scale implementations, reducing risk, and improving user acceptance.
• Reusable building blocks: Core modelling logic and solver controls can be adapted across PoCs, speeding development of new solutions.
• Stronger stakeholder engagement: Early visibility into working prototypes gets business users invested and excited about optimization’s potential.
• Steady pipeline of projects: Rapid PoC turnaround keeps new optimization ideas flowing across PepsiCo’s global operations.

Extending the Impact of Optimization at PepsiCo

PepsiCo’s success with daily production planning and headcount optimization is part of a much larger effort to embed optimization across the enterprise. What began as two targeted solutions has grown into a broad, ongoing partnership with FICO focused on building internal capability, earning user trust, and driving continuous innovation.

As PepsiCo’s use of FICO Xpress Optimization has expanded over the past 15+ years, the company has seen consistent performance gains from FICO’s commitment to continuous improvement, including:

• 15% faster solve times realized immediately after upgrading from the prior Xpress release 
• Significant performance gains achieved through solver tuning in collaboration with FICO’s support team
• New feature and performance enhancements delivered with each release

How FICO Can Help You Optimize Operations

• Read the full case study of Pepsico and FICO
• Discover the power of FICO optimization
• Explore FICO optimization on the FICO Community