Development and implementation of optimization sales and operations planning and management system (S&OP) for MMK Metalurji
- Customer
- MMK Metalurji
- Project manager on the customer side
-
- IT Provider
- GMCS
- Year of project completion
- 2024
- Project timeline
- July, 2023 - October, 2024
- Project scope
- 30000 man-hours
- Goals
-
The key project goal is to model the entire supply chain of the enterprise in the system, from the procurement of raw materials to the delivery of finished products to the end consumer, covering two production sites of a metallurgical plant in Turkey.
The project’s economic purposes are:
- Increasing the company's marginal profit;
- Improving the level of customer service;
- Minimizing logistics and warehouse costs;
- Improving utilization of factory equipment;
- Providing the analytics using scenario and plan-fact analysis;
- Reducing risks through comprehensive planning at the level of the entire production chain of the enterprise.
- Project Results
-
Two business planning processes have been implemented:
- Optimization rolling planning with a weekly or monthly cycle and daily detailing. As a result, all involved departments receive guidelines for optimal activity for 3-4 months ahead.
- Budget planning with a monthly and annual cycle. Discussion and approval of the budget are carried out directly on the basis of system materials, and all budget certificates and reports are ready immediately after the calculations are completed.
Standard calculations take only a few minutes, allowing users to receive a package of necessary reports in real time. Using the system for monthly budgeting provides a reduction of labor costs for preparing and coordinating the budget by several dozen man-hours.
The system provides reporting both in standard tabular form (including regulatory reports) and in graphical form (including dynamic dashboards). In total, more than 20 reports are available.
The uniqueness of the project
The optimization planning system developed during the project has 2 distinctive features:
1. Multi-level planning
The system has three applications with different levels of detail and planning horizons:
- Monthly Planning (MP), 1-18 months
- Weekly Planning (WP), 4-16 weeks
- Daily Planning (DP), 7-28 days
First, the planning is performed with a longer horizon and/or with greater detail. Once approved, these plans become a reference point for the formation of more detailed daily plans. For example, the approved monthly budget is transferred from MP to DP application as a task for detailing by days. In the DP application, the daily plan will repeat the results of the MP monthly budget plan. In the end, all departments receive a detailed daily plan for their work, ensuring budget implementation.
At the same time, the data for DP can be updated to reflect the changing situation on the ground. For example, if there is a delay in the delivery of raw materials after the budget is approved, the DP model will produce a plan that is as close to the budget figures as possible with the new raw material availability in mind.
2. Rolling planning
Each subsequent planning cycle does not start from scratch. Some key data about the starting point of planning is automatically taken from the previous cycle in the system. For example, to calculate the budget for the next month, there is no need to forecast the initial stocks. Stocks can be taken from the up-to-date daily plan.- Used software
-
Software for the main components of the system:
- IBM Planning Analytics
- IBM CPLEX
- Microsoft SQL
Auxiliary software:
- IBM Cognos Analytics
- VMWare
- Docker, Hyper-V
- Difficulty of implementation
-
- Setting up new processes and automating previous supply chain planning processes in a single system
- An international project team, whose members spoke three languages and were located in six cities in two countries.
- Project Description
-
Initial stage of this kind of projects is associated with the creation of a database of reference data and the setup of processes to prepare data for planning. Since our company did not have systems of this class, the GMCS company together with our experts developed a methodology for preparing the reference data. Methods for aggregating and grouping stock items to form planning items were agreed upon.
Based on the methodology, a starting set of reference data was prepared, and a process for updating and preparing data for each planning cycle was designed. During the project implementation, it became clear that the detailing of the initial version of reference data was insufficient. Therefore, a more detailed version was developed, as close as possible to the nomenclature of the accounting system.Almost all departments and functional areas participate in the implementation and preparation of data for planning: the department of economics, sales, procurement, logistics, quality control, IT, and, of course, production departments.Key parameters of the system scale in terms of the data and reference data are:
- 30,000+ nomenclature planning items (raw materials, semi-finished products, waste, finished products)
- Dozens of suppliers of purchased items
- 17 production resources at 2 sites
- 100,000+ process maps, formulas and output statements
- 6 locations
- 3,000 clients, unlimited number of orders
The system models the entire supply chain of an enterprise, from purchasing raw materials to delivering finished products to the end consumer.A user can add hard and soft restrictions to any elements of the supply chain, which will be considered when forming an optimal plan. For example, you can set the maximum volume of purchases of a key type of raw material and see how its shortage affects all components of the plan.Additionally, the system can be configured to take other parameters into account. It can be required to ship the maximum volume of orders or guarantee a certain level of safety stock. Several target functions can be applied simultaneously, in a given sequence, or with coefficients of their mutual criticality.The full list of target functions implemented in the system includes:- Maximize Total Profit
- Minimize Total Expenses
- Maximize Amount of Satisfied Customer Orders
- Minimize Violation of Order Timetable
- Minimize Violation of Minimal Resource Load
- Minimize Violation of Maximal Resource Load
- Minimize Violation of Minimal Stock Levels
- Minimize Violation of Maximal Stock Levels
- Minimize Violation of Minimal Purchasing Amounts
- Minimize Violation of Maximal Purchasing Amounts
- Minimize Violation of Minimal Shipment Amounts
- Minimize Violation of Maximal Shipment Amounts
- Minimize Violation of Minimal Produced Amounts
- Minimize Violation of Maximal Produced Amounts
- Minimize Violation of Hot Complex Target Production
- Project geography
-
- Dörtyöl (Iskenderun)
- Dilovası (Istanbul)
- Additional presentations:
- MMK Metalurji_S&OP optimization.pdf