This company is one of the main players in the steel market and has been consolidating its business by constantly unlocking the potential of steel. With 80 years of existence and a large commercial network throughout the country, it has expanded its activity using innovative solutions and finding new opportunities with the objective of solidifying its market position.
They offer a wide range of products and services such as steel and alloys, cutting and industry tools, cutting and machinery services and thermal treatments.
The core business of the company, the cutting process, uses a combinatorial tool that allocates in the most optimised way possible, like a 3d puzzle with all the client’s pieces ‘inside’ one block of raw material.
This adds a great level of complexity to the sequencing process in the form of virtual orders. These orders can be characterised as being necessary to perform any client’s piece, however, the necessary block is not yet cut, so it is not possible to produce at the time of the sequencing.
This sequencing process was too bureaucratic, in such a way that every day, the new production orders were manually released to the plant floor without a clear vision of their impact on the factory load.
Due to the high volume and complexity of dealing with orders which had multiple dependency on virtual orders, there was no on-time visibility about their planning. This mismatched information led to no monitoring regarding plan execution, and therefore, less accurate data to manage client lead time.
To meet the proposed challenge, the implemented solution was based in a sequencing tool, capable of not only optimising the sequencing process, but also creating complete visibility of the state across all production cells and PO status.
By interpreting real-time data(PO, machine availability and release plan) from the client SAP, the algorithm manages to find the best solution which can be validated through a user-friendly interface before it is released back again to SAP.
This four-step approach is highlighted below:
The algorithm is able to face all the complexities regarding precedent order management and also deal with multiple machines for the same task with different cycle times.
The developed solution added to the production plant a series of new capabilities that promoted better efficiency, visibility and communication between departments.
The production orders were allocated more efficiently with better distribution between machines. Now, it was possible to interpret in real-time production planning, its accuracy, and the estimated time of completion.
From then on, it was possible to assign a well-defined sequence of operations, view load-capacity and production lead time and share all this analysis with several departments.
These projects highlight the amount of information that is available for people which isn’t documented as well as the large amount of data that exists in the systems and is not properly leveraged to generate actionable insights.