You are here
AI shows the way to efficient chemical production
Innovation and the use of technologies like Artificial Intelligence (AI), machine learning, IoT (Internet of Things), big data analytics and mobility are changing processes and business models across all industry verticals and the chemicals and petrochemicals industries are no exceptions.
In this area, Japan’s Mitsui Chemicals Inc is known globally for developing and delivering high quality chemical products that find applications in a multitude of areas including automobile, ophthalmic lens materials, healthcare, packaging and agrochemicals, among others.
In order to stay at the cutting edge and relevant, Mitsui Chemicals has been innovating by integrating Information and Communication Technology (ICT) into its workflow. “Today’s global manufacturers are innovating in their production processes and we also urgently needed to digitalise our production using IoT technologies and big data to effectively compete in the market,” a Mitsui Chemicals official says.
Even though Mitsui Chemicals was formally established in 1997, with the merger of two companies, it has a proud heritage that dates back to 1912. In that year, Mitsui Coal Mine started a chemical department.. At that time, coal chemistry were one of the cornerstones of Mitsui’s chemical operations till the transition to petrochemicals in 1960. After several companies and subsidiaries, Mitsui Chemicals was finally established 21 years ago.
The mission statement of the Mitsui Chemicals Group has been to expand its business domains of Mobility, which is mainly centred around materials used in the automotive industry; Health Care, which encompasses a variety of products including ophthalmic lens materials, dental materials, and nonwovens; Food & Packaging, which includes agrochemicals, packaging, and related materials; and Next Generation Businesses, which covers the development of solutions in an energy, agro-system, medical, IoT, and related fields.
The Mitsui Chemicals Group aims to achieve sales of two trillion yen and operating income of 200 billion yen by 2025 by expanding its key businesses of mobility, healthcare, food & packaging, development of next generation business and stable earnings in the basic materials business.In Singapore, the company’s presence includes Mitsui Phenols Singapore Pte Ltd, a phenol business as well as Mitsui Elastomers Singapore Pte Ltd, which manufactures and sells high performance elastomer "Tafmer" used as a resin modifier for automotive materials and packaging materials. It also operates Prime Evolue Singapore Pte Ltd which manufactures and sells polyethylene "Evolue". These manufacturing facilities are located on Jurong Island.
AI for growth
The company feels that the use of AI can help in creating more efficient production processes. It also wants to use AI to create a depository of knowledge as its workforce ages.
“As our operators age, we need to standardise and share their knowhow as ‘implicit rules and knowledge’ among our production team,” the Mitsui spokesman added.
Mitsui Chemicals has been dabbling with AI, especially neural networks, for more than 15 years. The resulting system is capable of analysing temperature, pressure and flow rates in the plant via its neural network. The goal is to improve production processes and raise product quality.
However, the bottleneck faced by the company has been the amount of labour required to generate the instructional data, which serves as a model for the neural network. It did not prove to be a cost-effective solution. In addition, the need to revise prediction models for every change in the production process was an obstacle to applying the system in actual use.
Recent advances in machine learning and AI gave the company hope that the processes could be automated and an AI could learn to solve real problems in a live environment. Mitsui Chemicals realised that it would have to rope in a partner in order to solve the challenges in producing a live AI platform. After considering various vendors, Mitsui Chemicals chose NTT Communications (NTT Com) as its partner.
Partnering with NTT Com
According to the Mitsui Chemicals spokesman, the company chose NTT Com because of its extensive expertise in AI technology which it felt offered “real advantages”.
“It was also refreshing to see the company eager to enter our industry despite the relative lack of application of AI among chemical manufacturers,” he added.
As a test case, Mitsui Chemicals launched a project along with NTT Com to utilise deep learning in the production of gas products. The goal was to predict the quality of chemical products based on sensor data from production processes, such as type of materials and reactor conditions.
Mitsui Chemicals has conducted numerous tests since the launch of the project using various prediction models. The concentration of gas after 20 minutes can now be predicted with high accuracy (average +/- 3 per cent deviation from actual concentration).
NTT Com was provided with 51 types of sensor data, by Mitsui Chemicals, including data on material temperature, pressure, flow rate, reactor set value and gas product concentration. The data was provided raw without any cleansing or explanation of the manufacturing processes with which they were associated with. Mitsui Chemical personnel were quite amazed at the accuracy of the predicted quality that NTT Com’s AI programme provided.
“We assume that such accurate predictions will enable us to extend our AI project to various other areas. For example, we hope to detect anomalies in sensors and measuring devices, as well as optimise maintenance schedules in the near future,” the Mitsui official adds.
Since malfunctions are extremely rare, the AI system will have to become familiar with normal operation so it can detect deviations from the norm. This is another area in which NTT Com is working with Mitsui Chemicals.
For the long-term, Mitsui Chemicals is aiming to create next-generation facilities that harmonise relations between humans and machines. For this, it will seek to implement the latest technologies and educate its process engineers on how to efficiently manage product data. “Business proposals from NTT Com will be a major asset in coming years,” the company spokesman adds.
Mitsui Chemical hopes that with the help of NTT Com it will finally be able to implement its deep learning technology with the AI learning to solve problems. “Since the conditions for implementing ICT have greatly changed over the years, it was about time to take on the challenge of introducing AI in our production processes,” the Mitsui Chemicals spokesman added.