Petronas- a global oil and gas company. Since its founding in 1974, it has business interests in 35 countries
Mobile employee concept
- When did you start implementing IoT technology in Petronas?
— In oil refineries, digitaltechnology has been around for quite some time. PLC (programmable logic controller - Hi-Tech), DCS (distributed control systems - Hi-Tech) are all also digital technologies. Various sensors that measure, for example, pressure - they existed before us. Our challenge now is to focus and put effort into improving our systems. Where we see problems, the IT team comes with the necessary software - for example, for predictions. With the help of incoming data, it is possible to build an early warning system and improve factories, therefore, be more competitive.
- Now you mainly use IoT technologies for predictions based on the collected data?
- There are two parts. The first is the automation of repetitive tasks and processes. When the operator checks the equipment, he does the routine work and writes all the data into his notebook. This data may be lost. We provide them with a digital platform to use laptops for entering data. All this data is immediately transferred to a centralized database. So the information is not lost, then it can be analyzed. Technicians write reports on portable devices, engineers analyze information, deal with necessary corrections. So we reduce the share of routine work and increase the efficiency of plants.
All information about spare parts, all problems -This is in portable devices. Employees can request the necessary details by their phone number. And the repair will be extremely fast. There you can even find the necessary instructions for repair procedures. We call it a mobile employee.
We also do predictive maintenance.Some equipment may fail, and there will be nothing critical about it. It is simply repaired, since it does not affect the operation of the entire plant. But you can’t do this with critical equipment; you need a strategy. Certain equipment must be turned off after 8 thousand hours of operation and inspected. But using predictive analytics, we can monitor the condition of equipment and extend its operating time by a certain number of hours. This way you can save a lot of money.
"The companies that do not have IoT, there is no data"
- How many factories do you have around the world?
- It is difficult to remember a specific figure, from 30 to 50.
- Plants differ from each other in terms of technology used?
- Some plants are far ahead, somefar behind. We are looking for best practices and transfer them to old plants. Now we have a single digital strategy. There are digital strategy leaders from IT and IoT - together they find solutions that will take the company to a new level. Based on this strategy, the head office will move towards the unity of digital systems.
- Do you have a single system for all data?
— Now we have a unified digital PI systemOSIsoft. This is the platform into which all data from DCS (distributed control systems - Hi-Tech) is loaded. Unfortunately, we do not have a uniform DCS standard. One plant might use a Honeywell system, a second might use a Yokogawa system, and a third might use a Foxboro system. There are a lot of them. But data from different DCSs enters a single platform. You can do whatever you want with this data - predictions, adding artificial intelligence.
- Do you use AI and machine learning?
- Yes. For example, we use AI in our purchases. With it, we analyze the available inventory in warehouses, this allows us to optimize storage. Because warehouses are a lot of money. But based on the data, we can say what we need to store, and what we should not, and in what quantity. If someone says that you need to store spare parts for insurance, we show the data.
Data is a tool that allowsstakeholders are better stewards of money. Those factories that don't have data have no way of convincing stakeholders to spend money at the right time on the right things, when it's needed. Companies that don't have IoT don't have data. If you don't have data, how will you convince stakeholders?
Sharul Rashid
For example, for a new plant we joined forceswith the Saudi company Aramco. This is a project in Pengerang, Malaysia. Production - 300 thousand barrels per day. We need to give them the data so they can see the effectiveness. If I were a stakeholder, I would not trust people; some distance is needed. I would need to look at the data and from it understand what is actually happening in the plant. I need it to be safe, effective, and data-driven.
"If the work does not require people, we remove them"
- You have factories around the world. Is it equally easy to introduce new technologies everywhere?
- It is necessary to understand the local culture, and therefore, to do it in such a way that everyone is comfortable. Of course, in South Africa, for example, it is more difficult. In Malaysia is easier.
- How many sensors on average are involved in one plant?
- The number of sensors depends on the size of the plant. At a large plant there may be 50 thousand devices, on a small one - 10 thousand.
- What are the main functions of the sensors?
- They are needed for control and monitoring. Sensors measure everything - pressure, levels, temperature, flow. All these parameters must not exceed certain limits.
- The introduction of such a number of sensors is quite expensive. How quickly do the costs pay off?
- When we want to introduce something, we spendPayback calculations for each project. It usually takes one or two years. Yes, it is expensive, but wireless communication gave us the opportunity to greatly reduce costs. Laying cables is very expensive, and today we can use wireless gas detectors.
- If we talk about ecology, does the data somehow reduce emissions?
— On environmental issues we work together withMalaysian Department of Environment. We have installed sensors that record emissions. Since Malaysia has launched the Renewable Energy Act, we need to comply with its requirements.
- Is it possible to strive for clean energy, working in the oil and gas industry?
“We need to try to find a balance.” I can't comment further on this.
- What will be the future of the industry? Will people stay at the factories at all?
- People are needed. But the number of people can be optimized. Petronas is reviewing traditional processes, and if people are needed in them, let them remain. If the work does not require people, we remove them. It depends on the specific process.