• Home
• Careers
• Intranet Access
• Contact

• Company
  Institutional
• Business
  Client Portfolio
• Cases
  Projects
• Partnerships
  AIMMS

Steel Production Cases

Tactical and Strategic Optimizer of the Supply Chain - AMT (ArcelorMittal Tubarão)

Based on Mixed Integer Programming and integrated with SAP APO® (SNP® module), a model was developed to optimize the Tactical and Strategic Supply Chain Plan at AMT. This module supports optimized and integrated calculations of production, distribution, strategic storage and sales considering the profitability per client as well as the logistic and production capabilities at AMT and its subsidiary Vega do Sul.

Problem Description

In august 2004, AMT (at the time CST, Arcelor Brasil S.A. unit) began its plan to expand its production capabilities from 5 million to 7,5 million tonnes per year, including slabs and rolled coils, through the construction of new production units, such as a new Blast Furnace, a new Coke plant (Coqueria Sol, 1,5 million tonnes of coke per year) and a new continuous casting machine.

The investment corresponding to nearly US$ 1,0 billion will allow AMT to reclaims its position in the slab international market - as of October 2002 part of its production is being used at a Hot Strip Mill plant at 2 millions tonnes per year.

At the beginning of 2006, after an analysis conducted by IBM Global Business Services®, AMT adopted SAP – APO® as its tactical and strategic planning tool. At the time, the key question to be answered by the new System were (a) how to sell the expanded capabilities at AMT, in an optimized way - in other words, which was the best production mix and (b) which clients (and at what level of service) should be selected so as to maximize the company profitability, given the capacity increase in several production areas.

The Challenge

AMT adopted APO® due to its native integration with the ERP SAP®, as well as its transactional functionalities such as "workflow" and "middleware" embedded. However, AMT discovered that some of its business requirements could not be represented within SNP® ("Supply Network Planning"), one of the APO® selected modules. Optimization of the raw material mix (liquid steel and scrap) in the converters and the selection of clients based on profitability criteria are examples of such functional deficiencies. Because of this, UniSoma was invited to develop an optimization component as an alternative to the native SNP® "engine" that meet the requirements of all the involved areas (Production Planning, Logistics and Sales) in planning process of the supply chain.

The Solution

A generic Mixed-Integer Programming model based on AIMMS language was developed to support 2 similar applications. Both were based on the supply chain at AMT, including units such as production plants, distribution centers, branches, clients and possible transport lines amongst them. The first use of the application (called "Step 0") was projected for the Sales Department at AMT. This is, in fact, a S&OP (Sales and Operations Planning) tool that helps the commercial area to establish an optimal client list, in terms of maximizing the contribution margin. The system considers constraints such as the production capabilities and the clients product mix - the model guarantees, for each selected client, that the product mix be completed, be them more or less profitable. "Step 1" was conceived for the Production Planning Department at AMT. It supports the generation of a Business Plan, according to monthly time buckets, of up to 5 year horizon.

The "Steps" are executed on user demand from the SNP®: the input data is transferred from the SNP® to the Optimizer through a volatile Oracle® database; in such a way that the results are sent back by the model to the SNP® through the same database. The interface between SNP® and AIMMS is done through XI® (SAP Exchange Infrastructure®), the NetWeaver® tool from SAP® for process integration. The technology integration between SAP-APO® and the Optimizer based on AIMMS was defined according to the following diagram:


As the diagram shows, the Optimizer is used as an external optimizing component (running on an independent optimization server). The data is transferred through a temporary Oracle® database. The "Call Message" is a Web Service Call which allows SAP-XI® to invoke the Optimizer. "A" is the Web Service that receives the call message and starts AIMMS through a COM Object ("B"). Once the optimization is complete, a notification message is sent back to the XI®. The solution is based on a combination of AIMMS Agent Control Technology with its Web Service functionality, allowing and guaranteeing simultaneous "Steps" 0 and 1.

Another important aspect of this integration project concerns the data model mapping from the SAP® and the Optimizer, in other words, the Data Integration Model. The Optimization data model was developed according to SNP® since the SNP® data model is closed. The majority of the input data from the Optimizer had a direct correspondence within the SNP® data model (that is, parameters with the same primary keys). Exceptions where treated through the customization of SNP® Z tables and planning books.

Customer Quote

"The project has been ongoing since May 2006 and in this period the UniSoma consultants developed and validated the operational prototypes for Portfolio and Business Planning simulation. Besides the optimized plans, the tool permits an analysis to identify opportunities in the variation of demand, prices and material purchase, assessing its impacts in production mix, the client portfolio and the level of service."
Marcia Schwarz de Assis - Project Manager - 10th of November, 2006

"The AIMMS based Optimizer developed by UniSoma is the 'heart' of the System and the most important component of the project. It optimizes demand based on customer profitability - a request which we did not find in any 'package' - indicating the most lucrative flat steel customers for CST and Vega do Sul."
Marcia Schwarz de Assis - Project Manager - 11th of April 2007

Unfoldings

The results generated in SCP Phase 1 (such as the production levels at the hot strip mills) are used as inputs for the capacity utilization planning at resources in the reduction (ex: Coke plants, Sinter mills, etc.) and utilities (ex: Oxygen plants, Calcining kiln’s, etc.) areas. At the time UniSoma was invited for this first project ArcelorMittal planned these areas using a customized Supply Chain Planning solution provided by PeopleSoft®, which, due to lack of optimization functionalities, did not consider (simultaneously and in an integrated way) all the resources and materials involved in the process. Due to this, end-users treated the problem in an iterative manner, generating isolated plans for each area, that is, generating plans for a critical area (regardless of synergetic interactions and only estimating material exchange between different areas), then using the obtained values to generate plans for another area, repeating the process until no more plans were required. This process consumed valuable analyzing time, making it impossible to simulate scenarios or, if needed, reprogram the current scenario, and did not identify economical opportunities.

In January 2009, ArcelorMittal Tubarão invited UniSoma to develop and deploy a mathematical model, integrated with SAP® and the Strategic and Tactical Optimizer, which could replace PeopleSoft’s® Supply Chain Planning module and generate the integrated and optimized balance of Reduction and Utility areas.

Besides generating optimized plans, the solution should identify opportunities for raw material purchase and surplus or by-product sales (such as electrical energy or tar).

About AMT

AMT is an Arcelor Brasil S.A. Company, together with ArcelorMittal Longos and Arcelor Mittal Vega is a world leader in the steel slab market, responsible for 12% of worldwide sales volume, and is one of the largest companies in Brazil. Strategically located in the city of Vitória, the capital of Espírito Santos State, in eastern Brazil, CST has established itself as a supplier of high-quality semi-finished steel products (slabs and hot-rolled coils) for the automotive industry.