ÇimSA used the IBM® CPLEX® Optimization Studio toolset to solve complex business optimization problems and deliver solutions faster. Deployed through a collaborative, one-team effort between IBM Lab Services, IBM Data Science Elite and IBM Business Partner Lumtify, the toolset helped reduce manufacturing and distribution costs by 3% and cut inventory costs by 10%.
During a period of worldwide expansion, ÇimSA Çimento Sanayi ve Ticaret A.Ş. needed a decision optimization solution to help manage its supply chain network and enable management to find opportunities to improve service while cutting costs.
ÇimSA used CPLEX Optimization Studio to run complex “what-if” scenarios quickly, optimizing business decisions to improve customer service and reduce cement manufacturing, warehousing and distribution costs.
As construction boomed around the world, the global cement industry grew at an average annual rate of 5.1% between 2003 and 2018. After China, India and the US, Turkey took its place as the world’s fourth largest cement producer. ÇimSA, one of Turkey’s largest cement manufacturers and a leader in the production of white cement, increased its annual output to over six million tons. In addition to meeting domestic demand for cement, ÇimSA also developed a global distribution network for export to Europe, the Middle East and North America.
With several kilns, grinding mills and silos in more than 10 locations and seven international distribution terminals, ÇimSA established a complex network of facilities. The company also incurred significant overhead costs related to supply chain management, cement manufacturing and transportation. To optimize operations—while factoring in fluctuations in currency exchange rates, taxes, demand and manufacturing capacity changes—ÇimSA managers increasingly used a legacy computing tool for sales and operations planning.
But as the company grew, the planning tool reached its limitations. “Our previous tool was not adequately integrated and could not input all the business dimensions we needed to consider,” says Nevzat Çakmak, Project Manager for ÇimSA. “We were limited to just a few scenarios and were not able to create all the ‘what-if’ scenarios we needed to support decision-making.”
“The previous system also had issues with slow solution computation,” says Sabri Suyunu, Analytics Professional and Co-Founder of IBM Business Partner Lumtify. “It would be 24 hours before ÇimSA could see results for an optimization problem. The old system was also unable to integrate all the data required to create accurate models.”
As ÇimSA continued to acquire new production facilities and expand into new markets, its top management team needed more speed and flexibility from its sales operations and planning tool. When Çakmak and his IT team started looking for more advanced optimization solutions, Suyunu shared his experience. “We had already used the IBM CPLEX Optimization Studio for eight years,” Suyunu says, “and I suggested it was because it was fast, easy to use and had the ability to solve all of ÇimSA’s complex problems related to optimization.”
The CPLEX Optimization Studio toolset uses data science and decision optimization technology to model and solve complex business optimization problems. It also enables users to develop and deploy optimization models quickly and create custom applications designed to improve business results.
After deciding to deploy the CPLEX Optimization Studio toolset, ÇimSA’s first step towards implementation was to acquire all the relevant business data necessary to develop accurate optimization models. “Preparing the data and ensuring data completeness was our biggest challenge,” says Suyunu. “We then developed a mathematical model through the advanced modeling techniques of CPLEX. After the model was written, we could start the fine-tuning and testing processes to solve large-scale supply chain problems which include millions of decision variables and constraints.”
For the first modeling project, IBM Lab Services, IBM Data Science Elite and Lumtify tackled one of ÇimSA’s most important optimization priorities: the design of its global supply chain network. “With so many different production facilities in different locations, you need to make decisions based on different ‘what-if’ scenarios,” says Çakmak. “These questions are not easy to solve manually or with a spreadsheet, but CPLEX can run through millions of solution variables to make supply chain optimization successful.” says Eray Çakici, Optimization Subject Matter Expert with the IBM Data Science Elite Team.
In addition to expenses related to transportation and electricity, the cost of burning fuel to heat cement kilns to 2,000 degrees Celsius represents one of the largest cost inputs in the cement industry. Determining whether to use coal, petroleum coke, natural gas or other fuels at a production facility and deciding on the supplier, the price and delivery schedule is an especially difficult and ongoing challenge. Which is why ÇimSA made fuel-mix optimization its second major CPLEX optimization project. “CPLEX answers key fuel-mix questions like when, from whom, the amount and the price.” says Çakmak. “There are big cost savings when you can get that right.”
One of the keys to ÇimSA’s successful implementation of CPLEX Optimization Studio is the custom-designed user interface. “IBM’s interface team was very talented,” says Suyunu. “They coded all the designs we created as soon as possible into user-friendly screens. Parameters and inputs can be managed on the screen, and then you can run the model again and see the new results on the screen.”
During the implementation, ÇimSA was impressed with the collaboration between Lumtify and IBM. “They are not like two different teams. They acted like they were a single team,” says Çakmak. “When we made a deep dive into any problem, they always responded quickly and fixed it.”
After ingesting data, building models, designing user interfaces and testing solutions, ÇimSA went live with the new CPLEX Optimization Studio platform in early 2020—just as the COVID-19 pandemic began disrupting business operations around the world.
“In the past, we had optimization planning meetings with upper-level management on a monthly basis,” says Çakmak. “But because of the rapidly changing planning environment caused by the pandemic, we now have these meetings once a week and this toolset has become much more important to support decision-making.”
In one use case example, the toolset provided management with critical information and perspectives about which production lines to close down during the pandemic and which production lines to keep open. “The system was designed to help management make informed decisions and to help them overcome the difficulties of this period,” says Suyunu. “In fact, the platform has become so useful that board meetings started to be held over this tool.”
“Now, we can foresee the probable outcomes of our decisions,” says Çakmak. “The CPLEX system is very effective in presenting the output in a very comprehensible way to our upper-level management.”
Even as ÇimSA continues to expand the use of the CPLEX platform beyond the initial use cases of global network design and fuel mix optimization, the company has realized significant business benefits, including:
- An estimated 3% increase in working capital turnover
- An estimated 10% reduction in inventory costs
- An estimated 3% reduction in manufacturing, warehousing and distribution costs
The speed at which CPLEX Optimization Studio solves problems and delivers scenario reports is also making the information more useful. Instead of waiting overnight for scenario report processing, the new system delivers results 12 times faster, typically producing output 10–15 minutes after pushing the “run” button.
“CPLEX has become an indispensable tool for ÇimSA,” says Suyunu. “It’s a tailor-made solution and has the flexibility to be added to and developed as required.” During the coming year, the company plans to expand the scope of the toolset into other vital areas of the business, including inventory planning.
“This is a very agile tool with the advantage of providing answers for any scenario,” says Çakmak. “It prepares us for the future by helping us make better decisions.”
Founded in 1972 and headquartered in Istanbul, Turkey, ÇimSA (link resides outside of ibm.com) is one of the world’s top three producers of white cement and exports to 65 markets in the Middle East, North Africa, Europe and North America. The company operates six plants in Turkey, one plant in Houston, Texas and is acquiring another plant in Spain. It ships from seven international terminals in Germany, Italy, Spain, Romania, Russia and the Turkish Republic of Northern Cyprus. A subsidiary of Sabanci Holding, ÇimSA earned TL 1.726 billion in sales revenue in 2019 and has almost 2,000 employees worldwide.
Based in Istanbul, Turkey, IBM Business Partner Lumtify (link resides outside of ibm.com) has provided IT consulting services for over 13 years to Turkish industrial firms, retailers, consumer product businesses, energy and natural resources companies, public service entities and financial service organizations. Areas of specialization include mathematical modeling, system optimization, analytics, data science, business intelligence, digital transformation, machine learning, AI and natural language processing.
Legal
© Copyright IBM Corporation 2020. IBM Corporation, IBM Global Business Services, New Orchard Road, Armonk, NY 10504
Produced in the United States of America, November 2020.
IBM, the IBM logo, ibm.com, and CPLEX 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 ibm.com/trademark.
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.