The Basics of MES
It’s important to understand the principles and technologies be-hind MES software, so it will work for your needs, rather than the other way around.
Manufacturing execution systems (MES) have been around for quite a while, and there are hundreds of choices out there.
As MES software covers a broad and often loose range of added-value topics, it is rare to find any two systems that offer the same scope of functionality, especially given the rapid progress of related IT technologies, such as the Industrial Internet of Things (IIoT), and initiatives such as the “Smart Factory.” It is therefore important to understand the principles and technologies behind MES software, so it will work for your needs, rather than the other way around.
At a minimum, MES software enables engineers and managers to follow the progress of production and related activities against a plan, and to modify that plan to reflect changes in customer demand, material availability and process capability. Some MES packages have advanced control and optimization built-in, covering all aspects of supporting operation management.
The best approach to MES software is to determine the key business goals that need to be achieved and then work backwards. This avoids endless wading through technical specifications and functionalities that can often appear confusing until the context of what is needed is clarified.
It’s much like researching options for a new car. If you stare at “cabin air ionizer” long enough, and read about how it works, eventually you will believe that you can’t live without ionization, even though you didn’t know it existed before you read the brochure. Similarly, many MES functions sound great, but might not be all that relevant for your immediate needs. This can help a great deal with the selection process, though engineers should also think about what their operations’ needs might be in the future.
The selection of MES software and supporting hardware technologies should pave the way towards a single platform that will meet long-term goals without the need for replacement. That’s because the cost of an MES system is not simply the cost of the software and hardware, but is mainly the investment in the changes that an operation will make to allow the MES system to work effectively. Each system effectively prescribes an overall “best practice” for your factory, the fundamentals of which, when compared to both the current and targeted operation, must be considered. The cost of change and cost of ownership can be more significant than the capital purchase.
MES and ERP
MES will sit together with other factory control systems. Almost every factory has enterprise resource planning (ERP) software, which in many cases appears to have overlapping functionality with MES software. However, ERP software is typically orientated towards logical planning for the factory, whereas MES software is more concerned with the physical operation. ERP software by itself does not have clear knowledge about individual manufacturing processes and capabilities, such that there is an intermediate process required to “translate” the ERP plan into something that can be executed; without MES software, this is most often done manually.
With MES software, requirements information from the ERP system is used as the basis to create specific production operations that the MES software will then manage. The MES software, in turn, provides information of greater accuracy and detail about operations back to the ERP system, which can be used to enhance planning next time around. The relationship between ERP and MES software is a key factor for consideration. Where is the point of exchange between the two systems? The simplest answer is to define the roles of the two systems according to their strengths. Most often, the planning is done in a simple and logical way within the ERP system, while the physical tasks that take place in production are managed by the MES in a live and detailed way.
The starting point when considering MES software is the end point of production: completed products ready for shipping. The critical metric, from the customer perspective, is on-time delivery. For the manufacturing operation, the critical metrics are the cost of manufacture and the capacity of the facility. Capacity is dictated not only by the range of equipment in place, but also how it is used when faced with the inevitable requirement of changing from one product model to another. The degree of flexibility of both automated and manual production operations dictates the overall effectiveness of the facility when making a high mix of products.
To determine the overall capability to meet delivery goals when many product types are being produced simultaneously is a difficult calculation. This is the first potential value of MES software. It creates visibility of the status and performance of processes at all times. It shows opportunities for improving production flow and even assigning work orders. Live tracking of products and subassemblies is a cornerstone of MES. Each production unit can be provided with a unique identifier, such as a label or etched bar code, so it can be tracked throughout the factory.
Simple MES Operation
The ID code of each production unit is read at each key station, providing several related values.
First, the MES processes information about each event, so the system can show the location of every production unit.
Then, using the timings of each reading, MES can visually display the flow of production units. Graphs can show key performance indicators to help production management and engineering understand the performance of the assembly line. For example, the software can identify bottleneck processes where products are queuing in front of a workstation. It can also find starving operations, where operators are idle as they wait for the next production unit to arrive. Production units that fail test or inspection processes and need rerouting to repair stations can also be seen, as can the disruption that it causes to the main production flow. Details about efficiency, productivity, overall equipment efficiency and other metrics can all be determined from analysis of data collected by the MES.
Tracking production units in this way leads to other benefits. Because each production unit has a unique identifier, the MES can ensure each assembly conforms to the work order. The software ensures that each assembly goes through each process in the correct sequence, even if the product is temporarily routed to a repair loop. This ensures that no production process, or more crucially, a test or inspection operation, is missed.
MES systems that are linked with engineering data preparation systems can ensure, for example, that operational standards, setup documentation and work instructions will match the product being made, including any revision variation or material substitutions. The ability to show such documentation to operators electronically eliminates opportunities for errors or omissions, particularly in audited situations where compliance is critical and revision control is everything.
Data from automated and semiautomated production processes, such as test results, can also be collected and attributed to each production unit, so that a history of production can be made for traceability purposes. The complexity of this depends on how much data can be obtained from machines and how much time is available for manual operator input. The machine connectivity and user interfaces are a critical part of the success of MES software. The ability to interpret data from many different machines, and the ability to gather manual operator input in a consistent and managed way, are clear differentiators of MES capability. With manual processes, MES interfaces can be customized to the operator rather than an operational location. Manufacturing operations often have dedicated physical processes for certain products—for example, functional test—so not all processes need to be manned simultaneously.
MES software provides fundamental visibility of the operation and knowledge about how it is working, while highlighting where improvements could be made. However, there are many underlying factors that need to be considered to work out how to approach any changes, and what effect those changes may have. Production depends on many other aspects of the entire factory operation. These include material preparation and logistics, quality management, engineering data preparation, and management of tools, resources and people.
MES can help with that, too. It can delve deeper into the aspects of factory management to include support for these dependent processes, so root causes of symptoms seen in the production flow can be addressed, and the dependent processes themselves can be optimized.
How and where does MES software work at managing dependent resources? Most MES software is designed for maximum reach across industries, supporting many different industry segments. In some cases, certain features are critical, and in others they’re almost irrelevant. There will probably never be a standard definition of what MES software must contain or what it is expected to include. Arguments on this subject continue to evolve as expectations change based on technology advances, such as the IIoT. Modern expectations are for MES software to connect bidirectionally with automated equipment, so as to reduce the amount of operator support needed and to reduce delays for data acquisition and process control. MES software should provide a single digital platform that connects these areas, so managers can easily see and resolve disruptions in production.
Today’s assembly plants are lean operations. Inventories of parts and finished goods are anathema. Raw materials and parts are delivered on a just-in-time basis. Schedule interruptions cannot be tolerated. Management must have up-to-the-minute information on material preparation, engineering data, and the status of key tools and resources, including people. MES software knows both the current and intended production plan, as well as the current progress of production.
The sequence for the provision of materials and utilization of resources can be predicted, allowing the MES to manage such resources in line with final production requirements, or even to adjust the final production schedule if resource capabilities are exceeded. The MES operates in real time, working in many areas simultaneously and with many different kinds of data. The range of information that MES accommodates depends the design and scope of the software. Basic systems will show production requirements for a specific product, and then monitor progress to show performance. More advanced systems will manage key aspects of the dependent areas, adding value to each in terms of automation, management and traceability.
Supply Chain Management
Materials are a critical element for successful production. The absence of even the most humble of parts, such as cables or resistors, can mean that production cannot be completed and perhaps should not even be started. The MES therefore needs to treat every part and material as critical. ERP recognizes materials simply by part number and quantity “on site,” which starts from arrival of the raw material through to work-order completion. This is often simply a multiplication of products made by the material counts in the bill of materials (BOM). Though locations of materials are supported in many ERP systems, material inventory by location cannot be relied upon, as it requires extensive manual material counting and data entry at busy times. Almost without exception, the ERP view of materials is quite different from the physical situation, due to spoilage and other unaccounted losses. As a result, ERP systems will frequently make poor planning decisions based on incorrect assumptions of material availability, which risks creating schedules that cannot be fulfilled.
The need for frequent, costly, disruptive stock-checks and resultant need for adjustments to the ERP system is a symptom of this issue. So, too, are instances when materials cannot even be found. A basic MES system can help by providing knowledge of material movements. More advanced MES systems will take full control of materials. This starts with the unique identification of materials, either individually for key components and subassemblies, or by carrier for bulk materials, such as reels of surface-mount parts in electronics manufacturing. The MES will then decide and allocate storage locations, creating and managing tasks for operators as they disperse materials throughout the factory.
Mobile terminals are an ideal way for MES functionality to be present wherever needed, as materials are scanned into or out of locations. This can help drive lean practices, such as kanban and JIT delivery. The MES knows the quantity of each material required for production. It has full visibility of the current and near-term production schedules, and it is getting feedback about material consumption directly from the production processes themselves. More advanced MES software will also collect spoilage data, so that near-perfect inventory control can be maintained. All this data can be shared back to ERP, enabling it to make better decisions. MES software can also manage advanced aspects of materials, such as storage environment requirements, baking cycles for moisture-sensitive materials, obsolescence and expiry, and material substitutions against the approved BOM.
MES software eliminates internal material shortages, so production is never disrupted for lack of materials. In addition, stock inventories can be reduced by 75 percent; material logistics can be reduced by 30 percent; and warehouse space can be reduced by 50 percent. For many, the potential savings in material-related costs alone is justification for purchasing advanced MES software.
Engineering Data Management
Part of the job of a modern MES is to display electronic documentation at assembly and production stations. As more processes become automated, such documentation is targeted more at the setup of the process, while the automation follows an assigned sequence of instructions. These instructions are typically formatted and optimized using software provided by the machine vendor.
The engineering data on which they are derived, however, comes from the design of the product and local BOM. The conversion of design and BOM data, and assignment of work to complete a product, is divided between the various automated and manual processes by the MES, depending on process capabilities and throughput metrics.
Without a data management tool, engineers must read-in design data from many different formats, confirm data consistency, make adjustments depending on local BOM changes, and then manually split the data out to the various production systems based on their know-how. This process typically takes many days in the case of complex manufacturing, such as electronics, not only for new product introductions but whenever revisions are made. With only a basic engineering data management process, a minor change in just one component can require the repeat of many steps and confirmations.
Crucially, this whole data management process effectively dictates that engineering decides the production configuration with which to make each product in advance, with very little flexibility. While this has been seen as a “cost of doing business” in the days of high volumes and low product mix, it’s not sustainable in today’s dynamic high-mix environment. With advanced MES software, the processes of creating a digital product model, taking data in electronic format from design and BOM files, and assigning work are automated following engineering preferences. The entire process takes minutes rather than days, whether for a new product introduction or a revision to an existing design. This is one reason why productivity often declines sharply as product mix increases.
The ability to effectively create process data on demand allows the MES planning function to decide the best configuration to use for each product, depending on the status of production and demand at the time. More flexible planning leads to significant improvement in asset utilization and productivity. The MES planning function will consider lead times for subassemblies throughout the product hierarchy, optimizing execution and considering all available configurations, changeover times, and product grouping strategies.
This lean approach to planning is executed for the coming hours or days, rather than weeks or months. This degree of flexibility allows a step-change in the ability of the modern factory to respond to changes in customer demand. Factories can smoothly make transitions across a high mix of products, maximizing utilization without the need to produce excess finished goods stock. As a result, factories have seen 20 to 50 percent improvements in productivity.
A variety of resources are required for any particular production process to operate correctly. If any of those resources are not in place, the process cannot be executed. Processes may also require a range of support devices, such as material feeders for a surface-mount machine. Feeders must be set up to support the parts they will feed, and the parts must be installed. There is, therefore, a requirement for the MES to include feeder preparation. Other examples include nutrunners that must be set to a specific torque value or test equipment that must be set up, calibrated and checked.
The condition and maintenance of resources is just as important as availability and setup. Duty cycles must be counted and managed, triggering routine maintenance.
A more complete MES can manage both of these tasks.
People as a Resource
One of the most complex resources to manage is the operator. Each operator has different skills, abilities and experience, such that they may or may not be eligible to work at certain production processes. Time off due to breaks, illness or vacations can impede a factory’s ability to meet production targets or even to start production at all. So, it’s essential to know that operators with the required skills will be available when each production job is run.
An MES can provide that information. More advanced MES packages can provide work instructions and assembly guidance to workers at the point of use. That means the workers at each station can be less specialized, and workers can quickly and easily move from one role to another, increasing flexibility.
Active Quality Management
Issues related to quality are always disruptive. Defects lead to needless inspection, repair, rework and retest cycles, with associated delays and costs. Even worse, whenever a defect is detected, it is likely not unique. Until some countermeasure is identified and applied, the defect is likely to recur. In some cases, it’s preferable to pause production at least until the nature of the defect is known.
In addition to routing and tracking defective products, the MES can gather electronic repair tickets from test and inspection processes, providing support for defect analysis and repair procedures. Disruption of production is minimized through the fastest possible defect analysis supported by the complete history of production of the defective unit, including both material and process events. Statistics can be used to find the unique set of circumstances that led to the defect occurring, and identify any other production units that have been made in the same condition.
For every production unit, the MES provides complete traceability information for compliance and conformance, ensuring that everything needed for a particular product is in place and set up correctly. In this way, an advanced MES provides active quality management, which helps minimize the cost of poor quality, both within the factory and in the market.
The most advanced MES software goes one step further. Whenever the cause of a defect has been determined, the circumstances that allowed the defect to occur should be managed out of the operation. MES software with a built-in failure reporting, analysis, and corrective action system can make that happen, leading to true zero-defect operation.
Equipment breakdowns are another serious issue for manufacturing. To avoid these unexpected events, regular maintenance must be applied to all critical equipment. The challenge, however, is to understand what maintenance tasks need to be carried out and how often. After all, downtime for maintenance is still downtime. Ideally, the right amount of maintenance should be done at just the right time.
With an MES, engineers can create a more advanced maintenance strategy, utilizing information about the accumulated work performed by each production process. Preventative maintenance programs are created that minimize maintenance to just what is required. This strategy can be applied to frequent maintenance tasks, such as cleaning and lubrication, as well as major maintenance jobs, such as motor replacement.
MES manages maintenance resources and coordinates the timing of maintenance tasks. For example, maintenance tasks can be done when the machine is idle for some other reason, or they can be scheduled as part of some overall planning activity. MES
maintenance terminals are mobile by nature, a key tool for the maintenance engineer. Using a mobile terminal, maintenance technicians can interact with the MES to learn the location and nature of the work, record adjustments and settings, and understand maintenance procedures.
Breadth of MES Choices
The range and breadth of MES packages is somewhat diverse. However, it is roughly possible to classify MES software into types.
Simpler, more generic, MES systems can be thought of as one type. They are targeted to support a wide range of industries with basic functionality and limited flexibility or customization. These systems can be useful, though they tend to be limited simply to automate the current production operation.
On the opposite end of the spectrum are MES packages that offer in-depth support of complex requirements, usually for a specific industry. Often
requiring extensive customization, this type of MES can be very expensive, including on-going support costs.
The “sweet spot” for MES software is something between these extremes. A modern MES system will work using established digital technologies, including the latest IIoT standards, making data collection from a range of automated processes possible with minimal cost of ownership. The modules will offer reasonably deep detail into key areas, following standard digital process modeling, with minimal need for customization.
MES systems of this caliber are quite configurable, and bring with them the advantage of providing “best practice” procedures in the way that production, engineering and other processes work. Such systems offer an achievable return on investment.
The smarter MES systems become, the more they are able to make use of the data they collect, and to create value within their operation.
Simpler, traditional MES packages lack accuracy and depth of detail, requiring frequent human intervention to keep the operation running. Such limitations mean that even simple operator target dashboards quickly become unused. Such dashboards are worthless if they display unreliable data.
The latest digital MES systems provide better decision-making and visibility. Having information in the form of qualified “big data” enables more advanced values to be created in the form of analytic processing. Besides providing highly accurate operational dashboards, advanced analysis of manufacturing data can reveal patterns, opportunities and strengths. Management can make decisions that generate operational and financial improvements that otherwise would not even have been conceived. This trend towards advanced levels of visibility will continue, even as MES software evolves to include artificial intelligence. This is a key differentiator between MES systems.