Production scheduling at scale is hard. For a global CPG snack foods manufacturer operating multiple continuous production lines across a diverse product portfolio, it meant hours of manual Excel manipulation, tribal knowledge baked into copy-paste workflows, and a single scenario run before the day was over.
A leading snack foods manufacturer faced critical operational inefficiencies in their production planning processes. The company operated two primary manufacturing lines with approximately 30 packaging units in a fully continuous process with no intermediate storage. This required precise synchronization across the entire system, where small scheduling decisions created significant operational impacts throughout the manufacturing network.
Planners typically spent more than two hours building a single production schedule, requiring data consolidation from multiple systems, manual reconciliation of information, and extensive communication of changes across teams. This manual and time-consuming process severely constrained the organization’s ability to respond dynamically to market demands and operational disruptions.
The manufacturing environment presented additional complexity through stringent operational constraints including contamination prevention requirements that allowed only one product at a time, sequence dependency considerations, and minimum and maximum run limitations. These constraints created a highly sensitive scheduling environment where optimization required sophisticated analytical capabilities beyond traditional spreadsheet-based approaches.
Accenture partnered with the manufacturer to build a simulation-driven scheduling platform using Simio — starting with a Python ETL embedded in a local Simio model, and evolving into a fully cloud-native deployment on Simio Portal backed by Azure Blob Storage, event-driven automation, and Power BI reporting.
The solution was structured around three main implementation blocks designed to address the complexity of continuous manufacturing operations:
The team captured existing scheduling logic and translated planner knowledge into standardized rules that could be systematically applied across the organization. This approach ensured that tribal knowledge and expert insights were preserved while eliminating inconsistencies in manual decision-making processes.
Accenture built a comprehensive Simio model that executed planning through comparative logic, evaluated demand requirements, estimated production durations, and generated optimized batches while minimizing changeover requirements. The model operated at the packaging unit level, where actual planning decisions are executed, providing granular visibility into operational performance and constraint management.
The solution connected model outputs to Power BI dashboards for analyzing key performance indicators including demand fulfillment, equipment utilization, and changeover reduction metrics. This integration provided planners and management with real-time visibility into operational performance and enabled data-driven decision-making across the organization.
A critical advancement was the integration of Python ETL capabilities directly into the Simio model architecture. This innovation reduced external dependencies, improved system robustness, and enhanced maintainability while eliminating the need for separate data processing systems. The ETL importer utilized Simio’s data connector functionality, creating temporary folders automatically managed by the platform where processed files were prepared for simulation consumption.
The embedded Python script acted as a bridge between the model and ETL processes, utilizing model parameters as configuration inputs and eliminating the need for separate configuration management systems. This approach allowed the team to work independently of external integration teams while maintaining full control over data processing and model execution.
Phase 1: Manual Data Validation - Utilized fully manual data processes with standard Python ETL capabilities to rapidly validate scheduling logic and create a working prototype that demonstrated potential value.
Phase 2: Structured Data Integration - Introduced more structured processes with regular database connections, stabilizing the planning process while reducing manual dependencies.
Phase 3: Embedded ETL Integration - Integrated Python ETL capabilities directly into the Simio model architecture, eliminating external dependencies and improving system robustness.
Phase 4: Cloud-Native Enterprise Deployment - Achieved full cloud-based architecture with data stored in Azure Blob Storage and Power BI connected to centralized data sources, eliminating local dependencies and enabling complete automation.
The Simio-based production scheduling platform delivered dramatic operational improvements across multiple dimensions:
Scheduling time reduced from 2+ hours to 1 minute - a 99%+ reduction in manual planning time. Automated execution eliminated routine data manipulation tasks, allowing planners to redirect their focus to strategic analysis and value-added activities rather than data reconciliation.
Enhanced demand adherence through standardized scheduling logic eliminated the inconsistencies of manual planning. The solution maximized throughput rates in the complex continuous manufacturing environment while reducing planner bias by applying global scheduling guidelines consistently. Sophisticated constraint handling improved changeover optimization, directly impacting production efficiency.
Real-time scenario analysis now enables planners to rapidly evaluate the impact of demand changes, equipment maintenance requirements, or supply chain disruptions without manual recalculation of entire production schedules. Automated reconciliation with the latest operational information reduces scheduling conflicts and improves coordination across manufacturing teams. Enterprise-wide visibility through Power BI dashboards enables proactive decision-making, while risk-based planning capabilities provide realistic views of expected performance outcomes.
The modular, AI-extensible architecture supports continued innovation without requiring extensive system modifications. Cloud-native deployment facilitates expansion across multiple manufacturing facilities and product lines, enabling standardization of planning processes while accommodating local operational requirements. The embedded ETL approach enables rapid integration of additional data sources, establishing a foundation for predictive maintenance, demand forecasting, and automated optimization enhancements.
This implementation validates a critical question facing operations leaders today: “Could we do this at enterprise scale?” The answer is decisively yes. The partnership between Accenture and this global CPG manufacturer demonstrates how simulation technology can transform traditional manufacturing operations through a deliberate, phased approach that proves value at each step.
The team designed a modular, AI-extensible ETL architecture supporting both tactical and operational planning modes. By deliberately using a local CSV-based prototype to prove the logic before lifting to the cloud, they built organizational confidence while minimizing risk. An Azure Function triggered at run completion bridges the gap between simulation output and actionable business dashboards — all without middleware or manual handoffs.
For simulation engineers, this case study provides a replicable architecture that can be adapted across industries and manufacturing environments. For operations leaders, it demonstrates the tangible ROI of investing in modern simulation platforms that integrate seamlessly with existing enterprise systems while enabling rapid scenario analysis and data-driven decision-making.
The successful evolution from spreadsheet chaos to automated, cloud-native scheduling represents more than a technology implementation—it’s a fundamental shift in operational capability that positions manufacturers for sustained competitive advantage in an increasingly dynamic market environment. When planners can generate and analyze production schedules in minutes instead of hours, organizations unlock the agility required to thrive in modern manufacturing.