MES stands for Manufacturing Execution Systems, but it is much more than that. The name refers to the system of delivering orders to the production floor. i.e. moving information from the business system to the production system. Traditionally, systems that have been labelled MES move data from production to business systems, not the other way around. MES covers the technology required to move information between both systems in both directions.
Many people understand the basic concept of what information is needed where, but few understand the meaning of the data or the where data comes from (the source). The virtual space between the Production Floor and the Corporate Systems is a grey area.
As an example, these two very different worlds use a different language (See article Understanding IT-OT Speak). Mescon is a company that has worked in both environments and understands the grey area. We also understand that every business system is unique, as is every production control system. To integrate the systems together, you need expertise in the production and corporate systems, and MES experts to bridge the gap. MES fills in the gaps between what you do and don’t know about your business.
What Mescon refers to as MES is known by many other names. The title MES can be applied to sub systems, or a system covering many systems. Some of the names include:
The ideal MES sits in the background and acts as a data pipe between the systems or disparate data sources. The ideal MES pushes and pulls the data from one system to another. It is a transaction manager.
The reality is that the ERP system and the automation system, by design, do not hold all the information that the other system needs. Typically there is a need, due to inadequacy in the existing systems, for the MES system to have its own GUI (Graphical User Interface) and data repositories.
The functionality of every MES is unique, because business systems and production systems are unique.
If one of the functions of the MES to be implemented is to present the big data available from the plant floor as information to corporate decision makers, it is best to present the data within existing corporate systems, but this is usually not a viable option.
If one of the functions of the MES is to take customer orders and turn them into production work orders, there are usually considerations that a sales system just does not care about. For example the order will need to broken up into separate batches to suit line capacity, equipment and ingredient availability.
At Mescon, we have come across too many businesses where they have been burnt with previous attempts to integrate IT and OT. One of the technical reasons why MES projects fail is that functionality that should be executed in either the control system or the ERP system has been designed into the other system. Control systems should not attempt to make business decisions, and business systems should not make production decisions.
Many people promise they can work magic with the data, but few understand or can collect the data. A valuable MES system makes the data meaningful so that appropriate decisions can be made to benefit the company. It covers many different disciplines.
With our extensive experience, we know that every project and customer is different. MES projects often fail for non-technical reasons. Successful projects require a team approach, executive buy-in, getting beyond mere concepts and talk, a staged approach, open and honest communication, lack of personal agendas, a deep understanding and respect for the perspective of all end users, and going beyond the hype and language barriers and understanding real data.
Our recommended approach for implementing an MES system is shown in the diagram below.
This is a short meeting where MESCON and the client discuss basic requirements and agree to the next steps and the way forward. This stage is documented by a simple summary of the meeting and the proposed plans. The expectations of all parties are discussed and ratified.
The details of this stage vary depending on site resources. It can be a small discussion with site automation experts, or it can be a full audit of site control systems and equipment. The purpose is to tell the client what infrastructure changes, if any, are required before data can be collected. Most sites are data-ready, but significant infrastructure or technology changes may be required. The report, when fully completed, can vary from one page to several hundred pages, depending on the complexity of the plant and existing documentation.
This stage is conducted as a workshop, asking the client what they want to see. It sees the formulation of a dream list. Typical starting points are:
The MESCON task is to facilitate the conversation, to listen and compile the wish list.
Following up on the wish list workshop, this is a report that suggests possible solutions and budgets for further stages. Most customers do not know what is possible, the likely cost and financial benefits (ROI) of a quality plant information system. This report forms the basis of the way forward without significant expenditure.
Often referred to as a Functional Specification or Description, this details what the client will get before the following stages are commenced and the system implemented. Solid planning, thorough documentation and pre thought will avoid issues later. Changes can be made at this stage with minimal cost repercussions.
At this stage, on-site implementation begins and the data starts to be collected for subsequent reporting. The earlier that data collection commences, the more meaningful and verifiable the data will be. Often, all the data requirements are not fully realised, so some customers will take the approach of collecting as much data as possible. Although many modern historians have incredible collection rates, the historians, the collection media and the storage capacity do have limitations.
Accuracy and usability of data is important. Inaccurate data will only produce inaccurate results. The data needs to be verified against the real information. There are several analytical algorithms (e.g. Decision Trees, Naïve Bayes, Clustering, Neural Networks, Time Series, Linear Regressions, Logistic Regression) which, when applied correctly, can provide significant insight and forecasting of plant data.
Once data is available, the prototypes are reviewed and verified. The actual available data starts to take on meaning as it is analysed. Appropriate filters, exclusions, etc. are developed. Reports, data analysis and algorithms during development will often use simulated data and simulated scenarios. This allows them to accurately visualise scenarios and verify the data manipulation processes. Many of the scenarios will only rarely occur with real data.
This stage uses live data and real information is presented in the desired format, typically reports. This allows you, the customer, to start making decisions from the information provided. Data verification and accuracy is an important component of this stage. Mescon have experienced many customers expressing shock and disbelief at some of the figures that come out and question the data accuracy rather than the business or operational processes.
Meaningful information often not only answers questions, it will raise questions. Further detail is often required to take the analysis and information to new levels. This may mean collecting data from other items of equipment, to a finer detail or from other data sources, e.g. the weather or economic markets.
The detailed data may need to summarised, aggregated or analysed to reveal trends and predict future occurrences.
It is rare, even though a lot of thought and planning has gone into the project, that the finished products are perfect. As you quantify and understand your system more, you will ask different questions, and want to see data differently such as:
We had a client at this stage decide to move away an Excel based client reporting tool to static webpages.
You can find out much more about our past projects, successes, failures and opinions within our articles; please browse through them, comment on them, and ask questions. We are here to help you. Feel free to contact us
Get in touch with us and see how we can help
your business by moving data and information between the
factory floor and the corporate decision
makers.
CONTACT US