This article first appeared in Digital Edge, The Edge Malaysia Weekly on July 25, 2022 - July 31, 2022
Data is increasingly becoming a key component for organisations. Thus, ensuring data health — which supports timely and effective decision-making — is of utmost importance. Healthier data means better business decisions, as it allows for higher confidence in the insights provided.
Large organisations such as Petroliam Nasional Bhd (Petronas) have huge data landscapes. Therefore, maintaining data quality and standards to sustain the desired data health is prioritised, while also making available the organisation’s data to its customers to promote transparency.
Transparency of data allows for a culture of accountability to be built within the organisation for the user and the quality of data that is created.
“For us at Petronas, there is a specific emphasis that the data must be trusted and must meet the data guardrails that ensure the data is clean, valid and complete and meets the desired quality to aid analytics and business savvy decisions,” says Datin Habsah Nordin, head of enterprise data at Petronas.
The national oil and gas company has 325 subsidiaries and operates in over 50 countries around the globe, which lends to its complex and ever-growing data landscape across its integrated value chain.
Petronas has taken a holistic approach to data management by working with Talend, whose data health solution addresses and resolves numerous pain points, such as enabling data governance that allows for collaboration to improve data accessibility, accuracy and business relevance while supporting regulatory compliance through intelligent data lineage tracing and compliance tracking.
“The first challenge is to bring together all those disparate data sources into a single [platform]. A single platform with a different role-based interface that says there’s going to be different people within Petronas who will interact with the data, each of them with different needs,” says Stu Garrow, senior vice-president and general manager for Asia-Pacific at Talend.
Data that does not deliver outcomes to an organisation is of no value. “What it means is that data is created across boundaries and across the business value chain. The data in Petronas plays a crucial role to enable the digital oilfield to maximise value creation, while allowing increased operational efficiency and production optimisation,” says Habsah.
The digital twin effort, which is currently in motion, utilises relevant data from the physical plants to allow the monitoring of the current performance of the plants in question while allowing for the simulation of future performance based on potential scenarios and predictions — which are then further analysed on the most appropriate action to be taken based on the gathered data.
“People are using the digital twins to model optimisation before they go into the real world, and are able to get incredible insights that you would have never thought possible,” says Garrow.
“This relies on consistent consumption of the same data through their AI and machine learning models that are generated back to the original sources that have been used for decision-making.”
Accessing the data across the value chain allows for the identification of gaps and areas that require further collaboration, so that value can be amplified from a business perspective without any amendments to the physical plants or requiring structural or engineering expansion.
Working with the right data and having the right analytics to derive indispensable insights allows organisations to make valuable decisions, signalling an abundance of opportunity.
To advance carbon capture for sustainable practices, capturing the right data is essential, Habsah says.
“It’s no different than, say, what we are doing in the digital oilfield, and how we are looking to further improve our yield; the centrepiece of it is still from the data that matters with the insights we want to derive,” says Habsah.
A keen eye needs to be kept on proper data management practices, as without the right data health, tools such as artificial intelligence (AI) and machine learning can only help so much without accurate insights, due to data that cannot be trusted.
“By making data available and accessible, as well as utilising emerging technologies such as AI, machine learning and Internet of Things (IoT), many opportunities are provided to harness the value of data to bring forward the right optimal performance,” adds Habsah.
The same data could be used to flow in two different directions — whether by feeding it for decision-making for analytics or to machine learning models, the same data can be used to support multiple initiatives.
To build a sustainable model and infrastructure in the growing age of data, the use of automation is key as there are not enough data professionals in the world who are capable of handling the amount of data that comes through, with manual handling of data an unsustainable pursuit.
The use of automation dramatically reduces maintenance costs, by not using valuable human resources to perform acts of maintenance that do not deliver incremental outcomes.
“Sustainability is about solving problems with our valuable people, the most valuable resource we have, which is the people who can harness the data using automation. And once that is sustainable, it’s about dealing with data at scale,” says Garrow.
Data is created every single second; however, not all the data brings value to the organisation. That being said, business outcomes to be achieved need to be identified and backtracked using the data sets, with critical data elements being able to provide valuable insights.
Coming in many forms such as structured, semi-structured, unstructured and so forth, the complexity of data management is exacerbated by the shortage of talent among data professionals globally.
“We are making a deliberate move to upskill people not just within Petronas, but also provide opportunities to those trained in universities, associations or institutions to have working experience with Petronas in an apprenticeship programme that is being developed,” says Habsah.
Most people are not aware of the abundance of opportunities in the digital world and in the area of data and she hopes for more interest among Malaysians in becoming data practitioners.
Huge amounts of value can be garnered from the use of data. Yet, the journey begins with the organisations’ sharing of data, where data is made available and transparent before it is monetised.
As Petronas moves towards evolving into a data-driven organisation, the Petronas Data Liberalisation programme was sanctioned in February 2021. The programme is positioned to allow twin access of data within the company while being guided by regulatory requirements.
“This is an unprecedented milestone for us because it marks our data liberalisation journey at an enterprise level,” says Habsah.
“We are continually harnessing the power of data to enable the precise decision-making and generate higher value, and are now pioneering the journey to build a unified data platform, which we call the Enterprise Data Hub (EDH).”
EDH captures, curates and makes data available across the integrated value chain of Petronas, while allowing multiple business segments within the organisation to easily gather data, where it can be processed and made available for analytical requirements.
The prioritisation of data creates values such as higher productivity, production optimisation, better management of the environment, digital twins, on carbon emissions, as well as ensuring the ability to forecast by understanding trends that were seen in the past.
“We position EDH as data as a service, whereby we ingest and curate meaningful data into data products that can be used by many users for access and discoverability, allowing for a single place to get all the data from Petronas for analytics,” explains Habsah.
“EDH is designed and built with robust cloud-based technology that allows us to scale analysis at pace, where our legal entities and subsidiaries can now leverage EDH to discover value from the data that is converging into a single analytics platform.”
Evolving into a data-driven organisation marks a change in culture, where data is relied upon for decision-making, and by utilising tools such as machine learning and AI to bring actual insights.
“As we move forward, the bottleneck of hiding the complexity of something that is getting more complex and making it simpler is really changing, allowing the right people to have the right data,” says Garrow.
Save by subscribing to us for your print and/or digital copy.