About cookies on this site Our websites require some cookies to function properly (required). In addition, other cookies may be used with your consent to analyze site usage, improve the user experience and for advertising. For more information, please review your options. By visiting our website, you agree to our processing of information as described in IBM’sprivacy statement. To provide a smooth navigation, your cookie preferences will be shared across the IBM web domains listed here.
Blending breakthrough — optimizing ceramics clay
Accelerating development with prescriptive analytics
Whether used for traditional materials such as tile, pottery, abrasives, piping and sanitary ware or for new applications in electronics, 3D printing and dental products, ceramics production is growing rapidly.
Projected to be a USD 400 billion global industry by 2025, ceramics manufacturing uses clay as a raw material — a fine-grained natural soil which must be mined and blended to meet precise production standards.
ESAN Eczacıbaşı Industrial Raw Materials Company (ESAN) is one of Turkey’s largest producers of clay for ceramics manufacturing. The company also exports ceramics-quality clay to other countries. Because of changing manufacturing specifications, fluctuating market conditions, variations in the physical composition of mined clay, energy costs and other variables, ESAN planners must constantly develop new recipes for blending clay to deliver a feedstock that meets quality and cost goals.
For years, the development of clay-blending recipes required significant amounts of time and staff resources. “The 20 different types of clay materials excavated from the pits each have their own physical properties which must, in turn, be compared against 15 production parameters,” says Alper Şener, Data Analytics Head of IBM Business Partner Selco Consulting. “ESAN planners used spreadsheets to develop different mixing percentages and options. However, it could take a week for the planners to come up with four or five good blending options.”
Results
Data-driven decision-making
Decision optimization models reduce clay-blending process from days to seconds through prescriptive analytics and data science
Reduced Cost
Reduces cost of raw materials for ceramics processed in plant by 14% through optimized blending of clay
Another issue related to planning schedules was that by the time a new clay-blending recipe was ready, market conditions had changed. Planners had to go back to their spreadsheets and recalculate. Further delays occurred when recipes were tested in a pilot production run to confirm that the recipe met real-world quality and cost targets.
Like many mining and manufacturing companies looking for a competitive advantage, ESAN added data analytics capabilities to its IT department to support the digital transformation of its business. When looking for high-impact opportunities to demonstrate how data analytics could solve important business problems, ESAN and Selco focused on ESAN’s Bozüyük clay enrichment plant.
In operation since 1979, the Bozüyük facility offered several advantages for piloting the use of data analytics in a production environment. “The products produced at Bozüyük are more complex than materials processed at some of our other plants, but processes and logistics at this location were more favorable,” says Sinan Türkmen, Industrial Engineer and Data Analytics Manager for ESAN. “Once the plant was chosen, it was time to choose the right software to help manage the transformation, and Selco already had a head start in this area.”
When ESAN began evaluating prescriptive analytics software for clay-blending applications, it invited eight vendors to submit proof-of-concept proposals. “After the first iteration, the number of vendors decreased to three and Selco was one of them,” says Türkmen. “It was clear that they were experts in logistics, supply chain management, data science and math modeling.”
“Early on, we looked through their process and identified some issues with how ESAN formulated their math models,” says Şener. “ESAN soon realized that we could model their problems correctly and eliminate bottlenecks. They also saw how well our IBM ILOG CPLEX Optimization Studio software performed.”
ESAN’s IT department conducted its own research into the IBM® ILOG® CPLEX® Optimization Studio platform. “Compared to other optimization software, including the open-source software we looked at, we concluded that the IBM ILOG CPLEX solution could also solve our nonlinear problems more easily,” says Türkmen.
Prior to deployment, Selco and the IBM office in Turkey developed a software licensing plan that could prove beneficial to ESAN. “IBM worked really hard to provide the best price on the optimization software license,” says Şener. “Our client didn’t have a second thought about the price of the product and could focus on the quality of work that the product would provide.”
Soon after choosing the IBM ILOG CPLEX solution provided by Selco, ESAN conducted a preliminary test of a Selco-developed math model. “Selco gave us their first clay- blending recipe and we tried it in the laboratories,” says Türkmen. “Our research and development team told us that this blending recipe offered cost savings compared to other recipes. It also had a much shorter production lead time.”
In addition to deploying the IBM ILOG CPLEX solution for ESAN, Selco also developed a user interface that planners could use to create blending scenarios and manage data. “We set up application, database and web servers — a complete architecture with a front and back end,” says Şener. “IBM ILOG CPLEX software provides and supports us with the APIs needed to integrate the system and provide a seamless experience for our clients.”
Within a month after starting the optimization project, ESAN began to see significant improvements in terms of accelerated development of blending recipes and the number of recipe options generated. “With IBM ILOG CPLEX, we developed an optimization model for ESAN which runs in just 60 seconds or less to create the optimum clay-blending recipe with 50 - 150 alternative recipe suggestions for different efficiency and production time scenarios to maximize the production capacity of the plant,” says Şener. “We are really proud of that.”
In addition to reducing the clay-blending process cycle from days to seconds, the IBM ILOG CPLEX solution deployed by Selco at the ESAN Bozüyük plant has yielded significant business improvements including:
• A 14% reduction in the cost of raw materials processed by the plant through predictive analytics and optimization
• A 10% reduction in the cost of energy by optimizing processes that use less energy
“There are also some unquantifiable performance improvements that have come from using this optimization software,” says Türkmen. “For example, the system proves that we can actually produce materials that the R&D team once said were not possible, because the old recipes didn’t incorporate these physical or chemical properties. In the future, we can derive revenue from these new products.”
A long-term goal of piloting IBM ILOG CPLEX software at the Bozüyük plant is to roll out the solution to the nine other plants that ESAN operates in Turkey. “Because of the success we have seen with clay enrichment,” says Türkmen, “we plan to start using this optimization tool at our other facilities which have more production capacity.”
ESAN also plans to share what it learned at the Bozüyük plant with its parent organization, the Eczacıbaşı Holding group. “ESAN wants to start three projects this year with predictive and prescriptive analytics, and we think data analytics is something that can help the holding company,” says Türkmen. “The company now sees analytics as a decision support tool and is much more encouraged to take actions about it.”
“When we talk to our clients about how to invest in analytics, we tell them they should go with a strong company with a broad range of products,” says Şener. “The strategy of IBM, where they connect their different analytical components to IBM Cloud Pak for Data and Watson, supports the strategy we recommend for our clients. IBM ILOG CPLEX is also a scalable product where you can use different components when necessary and increase capabilities when you have more solutions.”
“As a data analytics manager, I was the only person in this role with the company,” says Türkmen. “Now we are enlarging this department and founding a Center of Excellence with data scientists. IBM and its ILOG CPLEX solution will be a tool for us in our projects and I am happy about that.”
About ESAN Eczacıbaşı Industrial Raw Materials Company
Founded in 1978 and based in Istanbul, ESAN (link resides outside of ibm.com) mines and processes clay and other raw materials for ceramics manufacturing. With 10 facilities in Turkey and over 2,500 employees, ESAN is one of the Turkish mining industry’s leading exporters. ESAN is a subsidiary of Eczacıbaşı Holding, one of Turkey’s largest industrial groups with a combined net turnover of TRY 11.1 billion in 2020.
About Selco Consulting
Based in Istanbul, Turkey, IBM Business Partner Selco Consulting (link resides outside of ibm.com) is a consulting company with extensive capabilities in supply chain and logistics management. By combining information technology and advanced analytics, Selco provides multidisciplinary expertise for supply chain consulting, business analytics, software consulting, material handling logistics and warehouse logistics.
Legal
© Copyright IBM Corporation 2021. IBM Corporation, Hybrid Cloud, New Orchard Road, Armonk, NY 10504
Produced in the United States of America, May 2021.
IBM, the IBM logo, ibm.com, CPLEX, IBM Cloud, IBM Watson, and ILOG are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the web at “Copyright and trademark information” at www.ibm.com/legal/copytrade.
This document is current as of the initial date of publication and may be changed by IBM at any time. Not all offerings are available in every country in which IBM operates.
The performance data and client examples cited are presented for illustrative purposes only. Actual performance results may vary depending on specific configurations and operating conditions. THE INFORMATION IN THIS DOCUMENT IS PROVIDED “AS IS” WITHOUT ANY WARRANTY, EXPRESS OR IMPLIED, INCLUDING WITHOUT ANY WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND ANY WARRANTY OR CONDITION OF NON-INFRINGEMENT. IBM products are warranted according to the terms and conditions of the agreements under which they are provided.
