A global manufacturing enterprise faced rising energy costs, fragmented visibility across plants, and limited intelligence in its legacy BMS. Akira deployed an AI-driven Energy Optimization Assistant using AWS IoT, Greengrass, and SageMaker to forecast demand, detect anomalies, and autonomously optimize HVAC and equipment loads.
The hybrid edge-cloud solution delivered a 28% reduction in energy consumption, 40% lower peak-demand charges, and significant cuts in manual interventions, achieving ROI within six months while enabling unified, real-time visibility and ESG-ready reporting.
Customer: Confidential Global Manufacturing Client
Industry: Industrial Manufacturing
Location: India, Middle East
Company Size: 4,500+ employees across 7 plants
The customer operated multiple manufacturing plants with energy-intensive equipment, HVAC systems, and complex load patterns. Due to rising tariffs, operational expansion, and dynamic production schedules, energy costs were increasing 12–17% annually. Their BMS offered limited intelligence, relying on manually configured rules that could not adapt to real-time variability such as weather conditions, occupancy, or machinery load.
No unified visibility across sites → fragmented BMS dashboards
High peak demand resulting in additional penalties
Overcooling/overheating in non-critical zones
No predictive insights on equipment inefficiencies
Manual control interventions leading to human errors
Difficulty meeting internal sustainability KPIs and ESG reporting
Lack of integration between energy data, production cycles, and shift patterns
Inefficient load distribution causing unnecessary operating costs
The customer needed an automated, scalable, AI-driven optimization solution that could intelligently balance cost, comfort, safety, and operational continuity.
The customer’s infrastructure had several limitations:
Legacy BMS systems without API-level access
No standardized data model across sensors, meters, and controllers
High data latency from edge devices
Inadequate storage for historical consumption and event logs
Non-uniform integration with PLCs and SCADA systems
Lack of forecasting capabilities for load and environmental dynamics
No automated anomaly detection for consumption spikes or equipment leakage
Security and compliance requirements added further constraints: encrypted data flows, role-based access, network segregation, and integration with the customer’s internal IAM.
Akira implemented its Energy Optimization Assistant—an autonomous AI agent that continuously monitors, predicts, and optimizes energy usage across sites. The agent ingests data from IoT sensors (temperature, humidity, occupancy), smart meters, BMS/PLC systems, and AWS IoT Core. It uses AWS SageMaker models to forecast demand, detect anomalies, and recommend (or autonomously trigger) optimization actions such as HVAC setpoint adjustments, load shifting, and equipment duty cycling.
The solution runs in a hybrid architecture: AWS cloud for analytics and model training, and AWS Greengrass-enabled edge nodes for low-latency control. Dashboards built using QuickSight unify visibility for energy managers, sustainability teams, and plant operators.
AWS IoT Core: Secure ingestion of sensor and meter data
AWS Greengrass: Edge-based execution of control logic and optimization actions
Amazon S3: Historical data storage + audit trails
Amazon Timestream: Real-time time-series queries for dashboards and insights
Amazon SageMaker: Model training for forecasting and anomaly detection
AWS Lambda: Event-driven triggers for agent decisions
Amazon CloudWatch: Monitoring and logging
Amazon QuickSight: Central dashboard for energy KPIs and ESG reporting
Real-time Processing: Sub-50ms edge inference with AWS Greengrass for autonomous control decisions
Multi-Agent AI: Coordinated forecasting, optimization, and anomaly detection agents
Scalable Architecture: Serverless auto-scaling handles millions of IoT events per second
Hybrid Edge-Cloud: Local processing for latency-critical actions, cloud for advanced analytics
The implementation followed a 10-week Agile delivery model.
Week 1–2: Discovery & Data Mapping
Audit of BMS, IoT devices, PLCs, and SCADA
Data schema standardization
Baseline energy performance modeling
Week 3–5: AWS IoT + Data Lake Integration
Setup of IoT Core rules and device onboarding
Pipeline creation using Kinesis → S3 → Timestream
Edge Greengrass deployment at plants
Week 6–8: ML Model Development
Load forecasting model built in SageMaker
Thermal comfort estimation model
Anomaly detection for energy leakage
Week 9–10: Automation + Dashboard
Integration with HVAC/BMS control loops
Unified energy optimization agent built using Lambda + Greengrass
QuickSight dashboards for ESG and operational KPIs
Testing, validation, and controlled rollout across zones
Security & compliance integration included IAM policies, encryption (KMS), network isolation (VPC endpoints), and audit logging.
The solution leveraged AWS best practices across the Well-Architected Framework:
Edge computing deployed for sub-second control responses
Continuous model retraining using SageMaker Pipelines
Secure IoT onboarding using X.509 certificates
Real-time optimization using event-driven Lambda agents
Scalable time-series analytics using Timestream
Automated deployment using CI/CD + Infrastructure as Code
Modular multi-agent architecture enabling future extensions
The project introduced unique innovations including hybrid ML inference, dynamic HVAC balancing based on occupancy prediction, and peak-demand prediction 45 minutes before occurrence.
The Energy Optimization Assistant delivered significant measurable impacts:
28% reduction in total energy consumption across 7 plants
40% reduction in peak-demand charges
34% improvement in equipment efficiency
55% reduction in manual interventions by facility operators
20% improvement in ESG reporting accuracy
Immediate ROI with payback in <6 months
The customer now has unified, real-time visibility into energy usage, equipment behavior, and zone-level performance across all facilities.
87% faster detection of consumption anomalies
Scalable data ingestion pipeline supporting 500k+ daily data points
Lower latency through edge-deployed inference
High availability architecture across multi-region setup
Improved data quality via Timestream-based time-series modeling
Strengthened security posture with fully encrypted data pathways
"Akira’s Energy Optimization Assistant has transformed how we manage energy across our plants. The AI-driven automation delivered immediate savings and deep visibility we never had before."
— Head of Operations, Leading Manufacturing Group
Integration with legacy BMS systems required custom OPC-UA bridges
Normalizing inconsistent sensor data formats
Managing hybrid cloud-edge deployment due to low-latency requirements
Fine-tuning predictive models for varying plant conditions
Always start with baseline energy profiling
Deploy edge inference early for smooth automation
Use time-series storage (Timestream) instead of RDS for performance
Build modular agent components for future scalability
Align dashboards with both Ops KPIs and ESG metrics
The customer plans to:
Extend the Energy Optimization Agent to two additional plants
Integrate solar forecasting for renewable optimization
Use Reinforcement Learning for autonomous multi-objective optimization
Expand the ESG module to include Scope 3 reporting
Implement predictive maintenance using vibration and thermal imaging data