People aren’t as clever as they think they are; and that includes you and me.
This is because of the all the information we don’t know we don’t know.
Computers are also not as clever as we think they are. Only a few years ago was one able to beat a human at chess. While anyone who has ever used Google translate will know that it could easily be renamed Gobbledygook translate.
This is because, up till now, computers have only ever known what we have told them. Artificial intelligence means that they are about to get a lot, lot smarter. Smart enough to analyse complex problems with more data than the human brain can handle, and yet reason an answer through logic as a human would.
This will have a profound impact on the chemical industry; a raw material sourcing, production, and sales industry packed full of data, processes, logistics and price margins that requires high-powered analytical thinking.
While still relatively new, AI is already being applied by a number of leading companies to make better chemical products.
One such company is the Tokyo-based firm Showa Denka (SDK), which by applying cutting-edge computing in their R&D projects has found a way to increase the efficiency of its polymer design operation.
As the company press release explains, “Researchers from Showa Denka (SDK), [in cooperation with the National Institute of Advanced Industrial Science and Technology (AIST), and the Research Association of High-Throughput Design and Development for Advanced Functional Materials (ADMAT)] conducted AI-based searches for polymers with desired properties. Aiming to demonstrate the effectiveness of AI technology in the process of polymer design, they focused on glass transition temperature, an index of heat resistance. Using 417 different types of structural data on polymers with known structures and glass transition temperatures, they conducted an AI-based search for a polymer with the highest glass transition temperature to see whether it is possible to shorten the development cycle.”
The result was that, “The researchers succeeded in discovering a target polymer with the highest glass transition temperature with an extremely small number of trials, namely, [an average of] 4.6 trials. This figure is about one-fortieth of the number of trials required under random selection of polymers, confirming the effectiveness of AI-based polymer design.”
The results surprised everyone, as the computer was able to predict an optimal polymer design for a specific task with fewer trials, in less time, and with less data than was previously thought possible.
Another project that has successfully applied AI for material development is based at Osaka University. This research focused on using AI computing for the automated selection of materials for organic photovoltaic (OPV) solar cells. These cells are made up of an organic component and a semiconducting polymer.
As the online industry journal Chemical Engineering explains, “The work sought to maximize the power-conversion efficiency (PCE) of OPV cells by determining the optimal combination of organic and polymer materials, a process that typically requires a great deal of time-consuming trial-and-error experimentation. Using AI and machine learning, the team was able to evaluate data from 1,200 different OPV cells to target the optimal set of properties — in this case, band gap, molecular weight and chemical structure — to quickly determine which ones would be most efficient, and then screen polymers for their predicted PCE.”
Once the AI had made its choices, the team was able to evaluate which of the resulting materials were most practical for manufacturing.
This particular form of AI application is called ‘random forest’ machine learning, as it requires the computer to form a network of decision trees for data classification and regression.
As the study’s co-author, author Akinori Saeki says, “Machine learning could hugely accelerate solar cell development, since it instantaneously predicts results that would take months in the lab. It’s not a straightforward replacement for the human factor – but it could provide crucial support when molecular designers have to choose which pathways to explore.”
Beyond the lab, AI is also helping chemical producers to optimise their current product ranges. By using AI to help consumers get the most out of their chemical products, Henkel has found a way to utilize high-powered computing and computer reasoning to aid chemical sales. To this end, Henkel employed AI for boosting hair product sales for its Schwarzkopf Professional range.
Dr Nils Daecke, Head of Digital Marketing at Henkel Beauty Care, explained how the system works, “With SalonLab, we are redefining the way both hairdressers and their clients are experiencing beauty in the hair salon. At the same time, we are laying the foundation for disruptive, data-driven business models, which build upon consumer insights and hair properties.”
As the Henkel website explains, “Around 10,000 hair samples were scanned, colored with different Schwarzkopf Professional products in different shades, and scanned again. Thanks to machine learning and the vast amount of scanned input, the computer is able to predict how the customer’s hair will look when a certain product is used. With augmented reality, the SalonLab Consultant App then brings the ‘output’ to life, showing the customer what he or she will look like after using a certain coloration.”
Smaller chemical companies may point out that an operation the size of Henkel can afford the resources to develop smart computing. Additionally, even if ‘off-the-shelf’ AI computing was affordable, then the time and manpower needed to collect sufficient data would still be prohibitive. It takes a lot of time to collect and process 10,000 hair samples.
This is a challenge highlighted by Tim Gudszend, global head of Adhesive Technologies and Investment at Henkel, when he noted that, “One of the biggest issues is generating all of the relevant data for a process and its influencing environment, and making this information available for a ‘big data’ solution.”
Certainly, at the beginning of any AI project, data collection may be a considerable undertaking. However, Dr Ata Zad, a polymer scientist and founder of the Canadian plastics company AxiPolymer, believes that while the amount of data needed to collect maybe daunting, it is the data’s size that makes using AI essential.
“When a company expands, the amount of new data it has to deal with grows exponentially — data relating to new clients, new formulations and products, new suppliers, new employees, and more — and it’s almost impossible to interpret these huge loads of data and get actionable intelligence from it with a traditional IT system.”
Adding that, “The only way to do this is with AI, which uses algorithms to find hidden patterns in data. These algorithms are iterative and will learn continually and seek optimized outcomes; they also iterate in milliseconds, which lets manufacturers find optimized outcomes in minutes instead of months.”
The true power of AI is its ability to handle so many data points. The chemical industry has a huge quantity of data which at present remains an untapped resource.
As Zad points out, “The main input for AI algorithms is historical data. In the polymer industry, the amount and variety of unprocessed data is incredibly high. It is clearly a matter of time until these tools become more commonly used to realize the full benefits of these data.”
If you would like to learn more about AI and its impact on the chemical industry, then you might enjoy reading ‘How Artificial Intelligence is Making Smarter Chemical Processes‘. Or take a look at other articles on the SPOTCHEMI blog page.
Photo credit: Japanesechemicalsdaily, Fticommunications, Henkel, Osaka University,& SDK