OVERVIEW
A paper manufacturing facility implemented an artificial intelligence-based energy orchestrator to coordinate power consumption across multiple production lines. By intelligently managing when equipment operates based on real-time energy pricing and production priorities, the company reduced total energy costs by 15-20% while avoiding power demand penalties.
INITIAL SITUATION
The paper mill operated multiple energy-intensive production lines simultaneously, dryers, pulpers, refiners, without coordination. Each line manager focused on meeting production targets without considering overall facility energy consumption patterns. This resulted in frequent power demand spikes when multiple lines ramped up simultaneously, triggering costly peak demand charges.
Energy represented a major cost driver in paper manufacturing, but the company lacked tools to optimize consumption patterns. Production planning happened independently of energy cost fluctuations throughout the day. The facility paid premium rates during peak hours while underutilizing capacity during low-cost periods. Additionally, power quality issues and grid instability caused by uncoordinated demand affected equipment longevity.
THE CHALLENGE
The objective was to reduce total energy costs while maintaining production output and meeting delivery commitments. The solution needed to coordinate multiple production lines, some with continuous processes that couldn’t be easily interrupted, while respecting operational constraints like shift schedules, material availability, and customer deadlines.
Technical complexity included integrating with diverse PLCs and control systems across the paper manufacturing facility, processing real-time data from all production lines (sampled every 1-5 seconds), and incorporating external variables like time-of-use electricity pricing. The artificial intelligence system required intelligence to balance competing priorities: minimizing energy costs versus maximizing throughput versus meeting delivery schedules.
THE SOLUTION
Mipu deployed NIA, her proprietary, AI-powered energy orchestration platform:
Phase 1 – Data Integration (6 weeks): Connected to power meters across all production lines, capturing consumption data every 1-5 seconds. Integrated with production scheduling systems to understand order priorities, shift patterns, and delivery constraints. Incorporated real-time energy pricing feeds including time-of-use rates and predicted price fluctuations.
Phase 2 – AI Model Development (8 weeks): Built machine learning models to predict energy demand for each production line based on product type, machine settings, and historical patterns specific to the paper manufacturing process. Developed optimization algorithms that simulate different production sequences to identify the most cost-effective scheduling while meeting all operational constraints.
Phase 3 – Orchestration Logic (12 weeks): Created intelligent decision rules that automatically coordinate when to start, slow down, or pause non-critical equipment during price spikes. The system respects production priorities, urgent orders always take precedence, while shifting discretionary loads to lower-cost periods.
Phase 4 – Validation & Control: Implemented simulation capabilities allowing operators to preview orchestration decisions before execution. Maintained manual override controls so plant managers retain authority during unexpected situations.
RESULTS
The energy orchestration solution transformed how the paper mill consumes electricity:
- 15-20% reduction in total energy costs achieved by shifting production to lower-cost time periods and eliminating simultaneous demand spikes
- Eliminated peak demand penalties by coordinating equipment startup sequences and avoiding the concurrent operation of energy-intensive processes
- Improved grid stability with more predictable, managed power consumption patterns reducing stress on internal electrical infrastructure
- Enhanced production planning with energy cost visibility integrated into scheduling decisions for the first time
The artificial intelligence system successfully balanced the complex optimization problem inherent in paper manufacturing: minimizing energy expenses while maximizing productivity. Plant managers gained unprecedented visibility into the relationship between production decisions and energy costs, enabling data-driven choices that improved both sustainability metrics and profitability. The solution proved particularly valuable during seasonal demand peaks when the paper industry faces both high energy prices and increased production volumes.
