You are Home   »   News   »   View Article

Maana CEO interview - analytics software company with investment Aramco, Shell

Thursday, May 12, 2016

Advanced analytics platform Maana has raised $26M in investment from Saudi Aramco and Shell, who join Chevron, GE, ConocoPhillips and Intel on the start-up's list of strategic investors. We interviewed the company CEO Babur Ozden

Silicon Valley advanced analytics platform Maana has raised $26M in funding led by Saudi Aramco Energy Ventures and accompanied by a new investment from Shell Technology Ventures. Existing investors GE Ventures, Chevron Technology Ventures, Intel Capital, and Frost Data Capital also participated in the current round. The company's total funding is now over $40M.

Many of Maana's investors are also customers, though the company has publicly confirmed GE and Chevron as customers.

Dhiraj Malkani, Investment Director at Saudi Aramco Energy Ventures and Andy Fligel, Director at Intel Capital, will join Maana's Board of Directors. Carl Stjernfeldt, North America Venture Manager at Shell, will join as a Board Observer.

Babur Ozden, CEO and co-founder of Maana, is particularly pleased that its investors are not ordinary 'financial' investors, but 'strategic' investors, looking to use the software as part of their company's operations.

'The business value that Maana is delivering to our customers is unlike any other technology available in the market today,' stated Mr Ozden, 'these companies are using Maana to rapidly optimize assets and business processes that no other technology has been able to resolve.'

'Maana is most useful for customers who want to analyse a large number of data sources, simultaneously, and optimize their assets and business processes,' Mr Ozden says.

Currently, many vendors do analytics with small data sets but struggle when they want to leverage disparate data sources at once, over a longer window of time, such as years of data, even decades when it comes to oil and gas industry.

Enterprise Knowledge Graph

Maana invented the 'enterprise knowledge graph' for industrial asset optimization. At the center of its advanced analytics platform is its patented enterprise knowledge graph. This breakthrough innovation in knowledge technology enables distributed, high-speed computation for data mining and machine learning across all data silos and data types to provide a holistic view of assets and business processes to be optimized. For example, data about a Well could be stored in as many as 20 different systems, but when someone is trying to understand the lifetime performance of a well they need to be able to leverage all that data.

This means that it does not need to move all of the data together into a 'data lake' in order to analyse it, as most other analytics technologies do. Maana takes a different approach, gathering data from multiple silos, and organizes it into a unique knowledge graph that brings together contextual information about an asset or business process. This knowledge graph enables analysts, subject-matter experts, and data scientists to truly collaborate -- for the first time -- on solving the toughest business problems. It also enables companies to dramatically accelerate creating hundreds of data models and quickly build line of business analytical applications using Maana's machine learning algorithms.

Domain experts can use Maana's platform to create their own data models using Maana and use that to explore questions they have about how data is related and correlated. Data scientists can use Python and Maana's API to develop custom machine learning algorithms on top of the knowledge graph.

In the past, to detect when a piece of equipment is going to fail, an engineer might need to use many different databases and data models, such as real time data models which record temperature and vibration, equipment maintenance databases, equipment failure root cause documents, even customer call records. This data may have been collected over years and is stored in many different systems. If you are looking at several pieces of equipment, over a longer time scale, it gets very hard to do, unless you have a way to bring all the sources into one knowledge structure.

Different employees in the company will want to have a different view on the data because they are trying to solve different business problems. Finance will want to understand costs associated with each well, drilling teams will want to understand drilling performance related to the well, HES teams will want to understand risks related to the well. The knowledge graph stores how all this data is related together so no matter which department wants to use the data they have access to the entire well knowledge structure. Without a knowledge structure doing cross domain analytics is very difficult and time consuming, this is where Maana brings a huge impact to enterprises.

The knowledge graph can be extended and enriched by using a mixture of machine learning (provided within the software), and guided by the company's business analysts and domain experts. For example if a oil company wanted to know which wells were 'complex wells' and have this complex well concept stored back in our Enterprise Knowledge Graph, an algorithm could be run on the data that looked at the well's geological characters (from the G&G data), the wells physical characters (from the Drilling data), and the wells production potential (from the production data). Maana's Enterprise Knowledge Graph can store this definition of complexity and compute a new piece of knowledge (complex well or not) from the underlying datasets and relate it back to the well. This new information is stored in the graph and related to all the other data that the well is related to enabling the different departments to ask questions about complex wells as they relate to drilling, G&G and production.

Case Study

Maana has several deployments at large global industrial and energy companies that have resulted in significant business value. An examples is at a company that manufactures and operates medical diagnostic equipment and wanted to reduce unnecessary parts ordered by field engineers. It knew that over 1/3 of parts ordered were unnecessary, which tied up $30m in cash that could be better invested elsewhere.

Maana used data from the company's XELUS inventory management and demand planning software, and SAP ERP. It found that 5% of engineers, located in the Asia Pacific region, had a 95% return rate on parts they ordered for maintenance and repairs. Maana worked out which parts would most likely be needed, based on past service issues where the correct space parts were ordered. This enabled the number of part ordered that went unused to go from 33% down to 11%, with savings in customer service trips, part returns, inventory and shipping costs.

Associated Companies

comments powered by Disqus


To attend our free events, receive our newsletter, and receive the free colour Digital Energy Journal.


Exploration Data Management : North Borneo Grid
Wong Teck Hing
from Sarawak Shell Berhad


Latest Edition Feb-March 2020
Feb 2020

Download latest and back issues


Learn more about supporting Digital Energy Journal