MaxinAI Energy

Data Science MAXimizing performance from Actionable Information

Our History

3+ years, on international market

  • 40+ projects, completed
  • USA, Germany, Israel, Switzerland, Poland

Group of 50 employees

  • Industry Experts
  • AI Engineers
  • Big Data Engineers
  • Software Engineers and Leads
  • 50 years of total expertise

- Organizers of DataFest Tbilisi, the biggest data conference in the region
- Data Science Georgia community leaders, organizers of School of AI

Our Team

Patrick

30 years energy industry

Served clients in 40 countries

From field engineer to business executive, technical expertise

Management positions in:
GE Energy, GE Oil and Gas, GE Power Conversion

Erekle

15 years in Software Engineering and Data Science

Ph.D. Data Scientist @ CERN

Director of Engineering @ MaxinAI

Founder and community leader of DevOps Georgia community, coach of AI startups

Ioseb

20 years in software engineering and data science

10 years in Govt. IT management

CTO of several startups

Background in Advanced Maths. Data Science expertise, organizer of Deep Learning Tbilisi

  • Industry experts
  • 10 Data Scientists
  • 8 Data engineers
  • 15 Back-end engineers
  • 5 Front-end / Web developers
  • 5 DevOps / Cloud engineers
  • 2 Quality Assurance/Testers
  • 4 Project Managers

Value We Create

Revenue increase:

  • Increased/optimized uptime
  • Increased/optimized utilization

Operations cost reduction:

  • Reduced downtime
  • Unplanned maintenance

Value delivered by:

  • Energy generation and consumption optimization
  • Predictive analytics models
  • Predictive maintenance models

Hardened by highly unique Data Analysis and AI model customization process

Our Approach

1.Exploratory phase

  • a. IBaseline customer KPIs

2.Proof of Concept Phase

  • a. Collect data from sensors
  • b. Collect data from different sources
  • c. Create Machine Learning Model
  • d. Validate Accuracy

3.Full-scale implementation

  • a. Integrate with existing systems, Deploy on Cloud, On-premise
  • b. Continuous Delivery, Continuous Improvement

2.a,b Data From Sensors, Different Sources

2.c,d Machine Learning Model, Validation

Create baseline Machine Learning model Use collected data to train Machine Learning model Validate model against validation data Improve model based on validation results Test against:

  • Historical data
  • Digital Twin
  • Real Scenario

Case Study 1

Entering Phase 3

Country: Germany
Industry: Energy transmission grid company
Problem: Grid Optimization
24 hours forecasting

Data:

  • Vertical Grid load values ~80 substations.
  • Environmental: temperature, humidity, solar irradiation, wind speed
  • Solar, Wind and Conventional power generation schedules and infeeds

Delivered:

  • Vertical Grid Load 24 hours ahead forecasting model with MAPE of 92%
  • Models for online Anomaly and Outlier detection/prediction online
  • Proposals for data collection and additional data aggregation process
  • Proposals for efficient application of open source Big Data management tools and applications

Case Study 2

Entering Phase 2.d

Country: US
Industry: HVAC systems
Problem: Reduce energy usage for ...
Decrease downtime

Data:

  • Inhouse temperature and humidity sensors data (up to 40 sensors)
  • Environmental: Outside temperature, humidity, wind speed and direction
  • Building occupancy data

Delivered:

  • Actor based model for keeping the optimal temperature, hence saving the energy consumed by HVAC systems, energy consumption monitoring dashboard.
  • Anomaly and outlier detection model used for HVAC predictive maintenance
  • Recommendations: data collection and aggregation methods with Big Data systems
  • 24 hours and highly accurate 1 hour ahead energy consumption by HVAC system

Case Study 3

Finished Phase 3

Country: US - California - Los Angeles
Industry: Real Estate
Problem:Short-Term-Rental compliance

Data:

  • Short-Term-Rental listing data from: Airbnb, Booking, Flipkey, HomeAway, etc.
  • Associated data, reviews, images, occupancy data and many more
  • US Government compliance data

Delivered:

  • Cross site Short-Term-Rental data identification and deduplication
  • Local Government compliance validation and checks, notification and letter management, CCPA compliance
  • Big Data solution for Analytical tools, near real-time data aggregation
  • Dashboard for Government Agency employees

Next Steps

Introduction, Question and Answer session

1. General Feedback

  • Detailed Presentations
  • Collaboration Details

2. Share Challenges, experience from similar projects, more case studies, real-life scenarios

3. Collaboration sessions

  • Workshop
  • Demo session

4.Exploratory phase kick-off

Please leave your email so that we can contact you later