Manufacturing Execution System: performance indicators

How a manufacturer who does not have the proper IT expertise tells a good MES from a great one if the user interfaces and functionality on offer are similar
(PresseBox) (Porto, ) MES applications are becoming more and more mainstream in manufacturing. They are being widely accepted as a key component of the IT infrastructure, along with the ERP, CRM and other enterprise level applications. As a larger number of manufacturers either adopt or consider adopting an MES application for their shop-floor, it becomes important to understand the parameters of MES performance. How does a manufacturer tell a good MES from a great one, as the user interfaces and functionality on offer might be similar for a lay person who does not quite have the IT expertise?

Ideally a mission critical application like the MES which may potentially change the way an organization works, should be chosen based on what it can do and how much better than the next application. In order to do so, generally a good approach is to have a team of operational, business and IT experts evaluate the application on all aspects, which should include, functionality/coverage, modularity/flexibility, scalability, overall performance and so on.

A project team working on MES selection would greatly benefit from a clear understanding of MES performance parameters and what makes an MES application more suitable for their operation. Today, we will do just that, provide an insight of what aspects of the MES application contribute to its performance and how evaluating these parameters can help choose the right MES for your production units.

An MES application at its very core is a transaction recording and reporting tool


Which means that any MES application should be able to record and report shop-floor transactions ranging from changing equipment states to movement or tracking of material/sub-material. To this end the most important parameter which defines the MES performance is the way in which the application captures and reports transactions and how fast it can do so. There is another catch here: the application should be able to perform considering the standard hard-ware available at most plant and not the high end stuff available at state-of-the-art laboratories.

The project team working on MES selection should first understand the process, material and general flow of the current operation to establish a sort of baseline performance. In case there are multiple production units, the standard should be the one which is most challenging, i.e. the plant that has the highest number of transactions, uses many material types and has complex flows, amounting to multiple transactions per second.

Then the team also needs to determine future needs in line with both business and operational requirements, predicting the increase that may be experienced in materials or equipment if the current plants are scaled up to accommodate business growth or to improve efficiency/throughput. It’s critical that the team recognizes the data flow needs both for the current and the future scaled up operation in order to set the requirements for the MES vendor clearly.

The speed at which transactions happen and the number of process elements to be tracked are important parameters in determining the required performance of the MES application. The complexity of an operation owing to high number of equipment, numerous material/sub-material inputs and sheer volume of transactions can render the best of MES applications incapable. Hence, the project team needs to get their potential application vendors to demonstrate how their MES is able to capture, stratify and report:
  • information; related to automated equipment state changes, ranging from one to many per second depending on the number of machines per line per plant,
  • engineering data collected say every half a second or less depending on importance of quality parameters and compliance,
  • Amaterial/sub-material hierarchy and the interplay between the equipment and material, orchestrated through the process coupled with high volumes.

The wow factor


In a complex process like semiconductor or electronic equipment manufacturing couple of hundred transactions might happen every second, along with necessary material tracking and reporting of engineering data, with each transaction resulting in one write and several read transactions on the database. This sheer volume of transactions renders many MES applications ineffective as they fail to accommodate the volume and complexity of transactions, making them slow, less flexible and able to provide limited and delayed material traceability.

In semiconductor or electronic equipment manufacturing owing to the automated nature of manufacturing and higher number of material types, the MES application responsible for executing the operation, should be able to capture transactions even in one hundredth of a second. In time critical environments, performance of the MES in terms of its speed and accuracy of transaction management becomes the key performance indicator.

Best-in-class applications can capture up-to 400 or more transactions a second with standard hardware, comprising 2 database servers, 4 application servers (2 virtual/2 physical), multiple client machines and a 1 GB network. The wow factor here is not the speed with which the best applications perform, but the fact that they can be scaled up or down, without affecting the functionality desired by the process owners.

Scaling up is possible from 100 transactions per second to 400 or more, by addition of standard application servers and maximizing the CPU usage, or increasing CPU or memory capacity, however the trick is to have the functionality of the software reside in a single layer, which enables seamless scaling as and when required. MES applications which are built and designed to be scaled are inherently capable of improving performance in a linear manner without reaching the diminishing returns area.

More about MES performance indicators in the next blog.

Download Test Report
: Affordable MES. Performance and Scalability for Time-Critical Industry Environments by Critical Manufacturing

Kontakt

Critical Manufacturing Deutschland GmbH
Maria-Reiche-Str. 1
D-01109 Dresden
Tom Bednarz
Managing Director / Geschäftsführer

Bilder

Social Media