Manufacturing Execution System (MES)


For operational production management

Manufacturing Execution System (MES)


The MES system enables:


  • Effective response to customer demand in real time
  • Continuous monitoring of supply signals
  • Maximizing customer satisfaction through fast order processing
  • Management of short-term and unpredictable product life cycles
  • Improving cooperation with partners
  • Decreasing inventory levels
  • Forecasting, modeling and optimization of resource utilization
  • Support for demand-driven production
  • Resource utilization decisions based on one consistent model
  • Capacity planning
  • Scheduling the placement of jobs to resources
  • Modeling the supply network process
  • Warehouse security level management
  • Adaptive production planning
  • Production planning according to the promised delivery date to the customer (Promise to Order Production Planning)
  • Increase in productivity
  • Increase in profitability
  • Gaining customer loyalty
  • Maximizing ROI
  • Transparency of the supply network
  • Optimization of capacity utilization
  • Shortening cycle times

Terminal for entering current information about the order.

MES and production preparation


MES (Manufacturing Execution System) as a tool for production management needs a complete definition of the product, while in ERP systems (Enterprise Resources Planning) the product is only partially defined. The logical conclusion is that it is possible to assign complete product data administration to the MES system. Using a transparent structure and appropriate software technology, the MES data defining the product is then available to all other applications.

The main part of the product data model is the technological procedure, which describes the production process of the product as a sequence of activities/processes with all required resources. As there are large technological and design differences between different products, there is a need to create an extensible universal data model:


The technological process determines what, how and with what resources will be produced.


  • Machines, equipment, jobs
  • Human Resources
  • Tools and mode of transport
  • Materials and semi-finished products
  • Production and control documentation and data
  • Definition of units
  • Order data


Data model for product definition:


  • Article: An article is a product that is produced using a technological process using various production resources. One manufacturing company can produce different items in different variants. Individual articles are then sorted into hierarchical groups.
  • Variant: It is possible to use one technological procedure, the same production operations and the same production resources for the production of individual variants of one article of the product.
  • Production operation: A production operation is a defined part of the production process which, as part of a technological procedure, is associated with a specific machine or device.
  • Technological procedure: The technological procedure defines the sequence of production operations and their assignment to machines and equipment. One technological procedure is applicable to the production of different variants of one article of the product. The actual time course of the individual operations of the technological process in the production of a single item of the product is not precisely determined until the time of adaptive planning of the order.
  • Product parts list: The product parts list contains defined individual parts of the product in the form of a hierarchical arrangement. These individual parts can be externally purchased parts or internally produced parts, which are also recorded in MES as individual product items.


Manufacturing process:


A manufacturing operation is an activity or process that is defined in the product development phase. It is independent of the place of execution such as a machine, device, or manual workplace. The relationship between the production operation and the machine, device, or manual workplace is defined only by the technological procedure. Material, production resources and information are needed to carry out a production operation.


Basic data of the production operation:


  • Title, description
  • Input quantity
  • The transformation that occurs during surgery
  • Output quantity
  • Time data of the production operation
  • Production time
  • Set-up time (quantity-independent)
  • Order Time = Setup Time + Production Time * Quantity
  • Human Resources
  • Number of employees for adjustment and production
  • Required skills
  • Used material
  • Direct material/Purchased products
  • Internally produced parts of the final product
  • Use of operational resources
  • Tools
  • Way of transportation
  • Packaging
  • Measuring equipment


Data model for resource management:


  • Description of the specific production background
  • Machines, equipment and workplaces
  • Basic data (name, serial number, images, etc.)
  • Production characteristics (maximum capacity, average setup time, production cycle time, machine hours, etc.)
  • Service data (warranty period, maintenance cycle, etc.)


Time models:


Time models allow the total time to be divided into production time and non-production time using exchangeability. For adaptive production planning, weekly or monthly shifts with defined working days are predetermined. In the world of a flexible labor market, it is necessary to apply these time models flexibly. In an extreme case, it is possible to assign different shift time models to one machine. Linking the time model with a specific machine, device or workplace then determines its production time, which has a major impact on job planning, time recording in general and also on estimating the capacity of machines, devices and workplaces. Therefore, maintaining the time model is very important for the overall quality of the data in the MES system. It is also important that changes can be made to the time model, in extreme cases even during the current shift.


Information and documents:


MES must transmit at the right time all the information that is needed in production to operate machines and equipment. This information mainly contains process and work instructions and additional documentation:


  • Work instructions (transmitted to terminals)
  • Construction
  • Manual setting of machine parameters
  • Monitoring functions, e.g. machine parameters by comparing target and actual values for statistical process control SPC (statistical process control)
  • Process instructions (passed directly to machines)
  • Automatic setting of machine parameters
  • Monitoring function
  • NC programs
  • Quality measurement plans
  • List of tolerances
  • Sample size
  • Details regarding measurement frequency
  • Description of the used measurement methods
  • Details on measures to be taken in the event of limit values being exceeded


Data from production terminals and sensors on machines in a timeline.

MES and production planning


The MES operational production control system includes the APS (Advanced Planning and Scheduling) advanced detailed production planning module, which performs collision-free, situation-based planning of custom filling with regard to limited production resources. The APS system guarantees that the execution of orders takes place in real time intervals and that possible deviations are immediately recognized through the continuous comparison of planned and actual values. The APS system contains adaptive planning algorithms (genetic algorithms working on the principle of artificial evolution and Darwinian principles of selection, mutation and crossover) for optimization based on realistic technological procedures.

The result of the planning algorithms is displayed as a Gantt chart for a better overview.

Scheduling algorithms take into account constraints from the customer by orienting the order of scheduled jobs according to the priority, importance of individual customers, or according to the minimization of internal costs. When planning orders, it is also necessary to take into account technological limitations (e.g. adjusting machines for individual types of products) and storage time and related storage costs, where in some cases the savings resulting from optimized machine adjustment are offset by additional storage costs.

Based on the product data model, which describes the product including its technological process and available production machines and equipment, adaptive production planning is carried out. Currently, the rule is that the data structure in ERP (Enterprise Resource Planning) systems relevant to adaptive production planning is very rough and must therefore be completed in the MES (Manufacturing Execution System). MES requires at least the following data from ERP for adaptive production planning (these data can be entered both manually and directly in MES):


  • Product label
  • Designation of machinery and equipment
  • Service
  • Technological procedures
  • BOMs


Based on the orders maintained by the ERP system, the MES creates production orders. The following data are required (these data can be entered both manually and directly in MES):


  • Unique customer order number
  • Unique product identification number
  • Quantity including unit
  • Date (earliest possible production date with regard to material availability)


One of the tasks of the MES is to break down the orders transferred from the ERP system into the optimal sequence using adaptive planning algorithms. What constitutes the optimal sequence depends on many frequently changing factors, however, two important aspects are needed when planning any production:


  • A view from the customer's side, in which the emphasis on delivery dates and quality comes first. The priority of individual customers is determined by the production planner.
  • A cost-oriented view to minimize production costs.


MES uses simulation to calculate and display different situations with regard to variants, quantities and deadlines. Fully automatic production planning is possible for production processes with few restrictive conditions and clear rules. In the case of more complex planning processes, the decision according to which variant will be produced is up to the management of the company.


Updated technological procedure - a condition for optimal production planning:


Years of experience with the analysis of technological procedures show that the entered values (mainly planned times for individual production operations, times for adjustment, cleaning, etc.) deviate significantly from reality. The reason for this is that the input of values takes place at a certain time before the start of production, based on estimates and time studies, and these values are then not refined during production itself.


Hence the important requirement for the MES adaptive planning function:

comparing time values from technological procedures and current production times and subsequent adjustment of technological times according to reality, which enables reliable and accurate production planning.


Recording of current production times is done simply and economically using terminals. Current times related to individual orders and products are recorded here. Statistically calculated average values of these actual times can then be compared with times in technological procedures and it is possible to analyze deviations.


Adaptive production planning must deal with the following complexity of assigned constraints, which are often at odds with the planning objective:


  • With regard to the contract:
    • Delivery date
  • With regard to the production process:
    • Minimum or maximum intervals between process steps
    • Transport times
    • Waiting times (cooling or aging processes)
  • With respect to production resources:
    • Resource allocation
    • Traffic availabilty
    • Cleaning and maintenance times
    • Availability of quality control resources (metrology, laboratories)


For adaptive planning of a larger number of orders in order to guarantee all delivery dates to customers, MES provides the following:


  • Synchronization of the process chain and minimization of unproductive and waiting times
  • Collision-free planning of custom filling


Strategies for adaptive production planning and planning algorithms:


The scheduling algorithm must be able to synchronize a complex process chain and perform collision-free scheduling of a large number of jobs while respecting the availability of production resources. Changes in quantity, terms, or exchangeability are entered manually, and the algorithm takes care of the rest. The planning result is then displayed, for example, as a Gantt chart. The planned data is then compared with the current production data.


Forward Planning/Backward Planning:


There are different strategies for optimization-based job scheduling. A suitable strategy should be selected according to specific conditions:


Forward planning:


MES will receive the earliest possible dates for the start of production based on confirmation of material availability from the MRP system. Scheduling is done using these dates. All production operations are planned so that they take place ahead of time. However, it is not always a suitable method. This is because if the final product is completed too soon, it can increase its storage costs and it is also possible that the material for production could have been used for a more urgent order. Therefore, if the delivery date for the customer is set in advance, or if it is only possible to deliver on certain dates, e.g. due to the delivered dates from the carrier, back-planning of orders is recommended.


Backward planning:


In this case, planning is carried out based on the final date of the production order and individual production operations are planned retrospectively.

Planned production order according to the technological procedure.

Zaplánovaná zakázková náplň s minimalizovanými prostoji mezi jednotlivými zakázkami.

Use of genetic algorithms for efficient planning.

MES and order processing


Various departments of the company participate directly or indirectly in the processing of the order. For each of these departments, MES provides a tool that guides workers in performing work tasks. In production, the machine operator receives a list of orders to be processed on the terminal. You can then view all the necessary additional information online. Other company departments that are linked to production, such as logistics or maintenance and repairs, also have the necessary information and a list of work tasks available from the MES. Production data are available for evaluation in a suitable format to the production manager, control department and company management. This evaluation is then the basis for the continuous improvement process. Access to data is available throughout the company (or worldwide) using Web technologies.


It is important for the machine operator to have a user interface (MES terminal) at his disposal that can be adjusted to exactly match the needs of a specific production process and to simply guide the user in performing work tasks. This is necessary for a successful MES implementation. Problems with the implementation of information technology (IT) in the manufacturing sector are very often caused by the non-inclusion or late inclusion of machine operators in the collection and configuration of information. The involvement of the worker means that the worker should first of all be informed about the goals and importance of the MES. Furthermore, in the second phase, the MES terminal should be developed and tested.


MES terminal




List of jobs assigned to each machine operator displayed in the order in which they are to be processed.

Each job is assigned a planned quantity, planned duration, planned non-productive time, planned start, and planned end. When the worker starts processing the order, the actual start is counted.


Collection and control of contract data


MES enables operational data acquisition functions (Operating Data Acquisition ODA)


  • adjustment/clamping
  • production
  • disconnection
  • cleaning


Each of these functions is activated only when needed for a specific operation.


Graphical presentation of the production plan


For an overall overview of the production plan, it is possible to display it in graphic form to the machine operator as needed.


Calling up information


If necessary, it is possible to call up information about completed orders, machine performance and the status of already completed production processes.

View events and exceed limits

If events such as machine failure or exceeding limit points (e.g. 6Sigma deviation) occur, they can be displayed to the user in real time with a link to the data.


Display of SPC/SQC data


This format is intended for displaying the results of statistical process control (Statistical process control SPC) or statistical quality control (Statistical quality control SQC)

For complex functions such as process instructions, test instructions, or assembly information, the worker can be guided by "wizards".

It is important to document all the states the machine is in (production, maintenance, error messages, etc.). Only in this way can the relevance of key indicators be guaranteed. Machine states are recorded automatically using sensors. However, it can happen that the machine gets into an automatically undefined state. In this case, the worker must select the cause of the machine failure from the list.

If the machine is in the state of production, it is necessary to record the produced quantity manually, or by a combination of manual and automatic entry.


Order preparation and adjustment


Tool change


After starting this function, the recording of tool clamping on the machine starts. Tool clamping is subject to many regulations that need to be displayed to the operator at the terminal during tool changes. Disassembly of the tools is carried out in a similar way. Here, the tool after disassembly is introduced into the tool store and the tool use time is added to the total accumulated use time. If the total usage time exceeds the specified limit, a tool preventive maintenance job is triggered, or at least a warning message appears.


Machine adjustment


It is a matter of choice whether to connect online machines within the MES. Modern technologies easily enable interactive communication between the MES and the control level. This communication is two-way. It means that the relevant (according to the order) data for the process can be uploaded to the machine and the machine status can be automatically fed back to the MES. In this way, after starting this function, control programs (e.g. Distributed Numerical Control DNC programs) are loaded into the machine, where they are then activated. It is also possible to start recording, displaying and checking data related to machine parameters. Several machines together can be connected to one terminal. The Supervisory Control and Data Acquisition (SCADA) system, which displays data in real time, can also be connected to the MES. Since in some cases it is necessary to adjust the machine manually, it is also necessary to display the adjustment parameters as well as the adjustment instructions of the machine operator to the terminal. This includes cleaning instructions. After the assembly adjustment is complete, the production function can be started on the machine.


Trial operation


After the appropriate tooling has been clamped on the machine and machine adjustments have been made, a test run can begin to test manufacturing tolerances, provided the material is available. At this stage, it is often necessary to make adjustments to NC (Numerical Control) programs. If the worker is not able to determine the production tolerances/quality himself, it is necessary to send the test samples to a metrology or laboratory. For this purpose, the MES terminal should enable a function that alerts the metrology or laboratory. The machine operator can start the production of the order only after the approval of the test sample.


Order control


Information management


Providing information is key to contract management. The first step is choosing the right job. This information is provided to the worker directly on the machine by the MES terminal as a list of new orders together with instructions for quality testing. It is also useful to provide workers with additional information such as target and current quantities during a production shift.


Inspection and tracking of manufactured pieces


To control the manufactured pieces, it is necessary to track them, i.e. to know exactly where they are, their condition, etc. at any moment in time. The concept of tracking describes how it is possible to retroactively reconstruct a specific production process, including all important events and data. It is thus possible to record all relevant details such as interrupts, messages, process values, parameters, etc. This collected data is archived and can be viewed retrospectively.


Order processing and production data recording


Operating Data Acquisition (ODA) is a group of functions for recording and checking all production data and includes both preparatory processes (e.g. adjustment, tool change, cleaning) and directly added value processes.

The "production" function associated with the process directly creating value is a classic area of ODA with the main task of documenting who, when, where, what, how long and how much is produced. The efficiency of the process with respect to quantity and time is recorded. This main function can be extended to other functions as needed:


  • Recording of machine downtime (e.g. downtime, reasons for downtime) with activation of maintenance (e.g. retroactive maintenance, preventive maintenance)
  • Material flow control
  • Process/product quality control
  • Control of waste material and rework
  • Performance analysis including cost control


Order processing time is determined by the adjustment and production function. The need for rework is captured directly in the process and a rework order is also created here, or the need for rework is captured by the quality control department, which decides on scrapping and generates a new rework order. It is important that these contingencies are subsequently included in job rescheduling, from which it is clear to which originally planned job the rework job is related, which is then scheduled as a planned job with all recording and control functions.


Process and quality control




After the initial boom of quality control management systems at the end of the 80s, they are currently undergoing a renaissance partly with new concepts. However, the essential here is the inclusion of quality control in production control. Quality control can be divided into statistical process control (SPC) and statistical product quality control (SQC). In fact, it is always a matter of keeping the production process stable by means of systematic methods in order to guarantee the statistical stability of the defined control parameters. The limit values of these parameters are either supplied by the customer or determined internally. Statistical tools should run automatically in the background, at least for process control, and in case of statistically unacceptable deviations, the SPC system should independently make decisions (e.g. about scrapping parts). The integrated quality management system is a key function of the MES and its task is to carry out continuous control. This forms the basis for concepts such as Define, Measure, Analyze, Improve and Control (DMAIC) and 6Sigma.


Process control


Target values for machine inspection are set during setup. During production, data is continuously recorded and checked, which is compared with the target values of the control parameters. Like the inclusion of statistical tools, this requires a well-thought-out data structure, both at the real-time machine level and at the MES level where data is stored and analyzed in a format that represents a compromise between data volume and relevant display.


Quality control


Even if a quality SPC statistical process control system is implemented as part of MES, it is necessary to control the quality of finished products using random sampling. Random samples should be taken according to the test frequency given by the inspection plan (eg every hour) as part of the SQC system of statistical quality control. Statistically evaluated data should then be clearly displayed in graphic form as histograms, etc. Here it is important to carry out these displays for each production section separately and if the limit values are exceeded, the SQC system should start the execution of corrective measures, which are retroactively confirmed electronically.


Performance data


Corporate Department


The following company departments are affected by reports from production


  • Production management
  • Corporate management
  • Controlling
  • Sales and Marketing
  • Process control
  • Quality management
  • Maintenance
  • Material management
  • Logistics


Production management


Production management has the main control function and is responsible for all production performance functions, therefore all deviations from the specified parameters are monitored here, and the production manager needs information and a real-time control view of production. It is necessary to display all unacceptable deviations from the target values so that the production manager can implement the necessary measures and inform the management of the company in the event of major problems. Here is a convenient information panel. From it, it is possible to enter the production structure, where the status of individual production equipment can be seen and individual production events can also be displayed as follows:


  • Process reports (measured parameters)
  • Quality reports (test parameters)
  • Maintenance reports (exceeding device usage limits and resulting maintenance times)
  • Reports on material storage (security levels)
  • Cost control reports (cost overruns)
  • Reports on the control of time courses (exceeding time limits)


Corporate management


Company management is primarily interested in whether the planned costs of individual orders and their delivery dates are met.




Kontrolling focuses on controlling costs in real time. Only cost variances are displayed here, and the controlling employee can check the reasons for the variances.


Sales and Marketing


The sales and marketing department receives MES reports from the adaptive planning of orders and their processing. With them, the sales and marketing department can continuously confirm delivery dates to the customer.


Material management


Material management needs to receive warning messages about requests for additional material supply from adaptive production planning or in the event of a drop in stock safety levels.


Process control


The process control department is mainly interested in unstable processes or critical machine situations. If the limit values are exceeded, the person in charge of process control can look at the types of individual events to take appropriate measures.


Quality management


If the limit values of the process and test parameters are exceeded, these events are reported to the quality control department.




The maintenance department receives maintenance reports and receives messages in case the limit values of the use of individual machines and equipment are exceeded. Maintenance dates for machinery and equipment are shown in the maintenance schedule. In addition, the maintenance department receives reports on breakdowns and downtime of machines and equipment as well as on received repairs (backward maintenance).




All reports related to material flows are directed to the logistics department

Key parameters and performance record

In addition to job analysis, it is important to record machine/equipment performance during defined periods and continuously monitor deviations.


Subsequently, the key parameters for evaluating the performance of machines and equipment (KPI Key Performance Indicator) are defined:


  • Availability
  • Grade of quality
  • OEE Overall Equipment Effectiveness
  • Performance level
  • Productivity time etc.


Manufacturing standards that have emerged over the past two decades are aimed at ensuring that every company must document its performance measurement process so that individual jobs can be traced back. Here again, it forms a basic element of MES with its wide range of monitoring processes.


Analysis and evaluation over longer periods of time




In the analysis of performance over longer periods of time, the focus is on recognizing weak points and the possibility of optimizing the production of individual products, or individual machines and equipment. For all products and machines and equipment, it must be possible to compare individual time periods with each other and thus recognize individual trends. This means that, for example, it is then possible to answer the question of which production department had the best overall equipment effectiveness (OEE Overall Equipment Effectiveness) over the last quarter. It must also be possible to evaluate statistical data, where a number of tools are available, from simple graphical displays to sophisticated tools for online processing (OLAP Online Analytical Processing), including multivariate statistical analysis.


Order efficiency


Cost analysis is fundamental to the analysis of the effectiveness of the contract. Accrued costs are checked in real time with respect to the work process and are continuously displayed so that timely intervention can be made. After the completion of the contract, this data should be available to the controlling department for the implementation of appropriate measures


Performance of machines and equipment with regard to the order


Important links can arise between the performance of machinery and equipment and individual products. For example, it is possible to recognize that machine availability for a certain product is always very poor, or that another product causes tool damage more often. In this way, it is possible to analyze the causes of deviations in performance in much more detail.


Dependency analysis


In the future, there will be a greater application of multivariate statistical analysis, which will make it possible to record dependencies between process parameters. Newer methods such as PAA Part Average Analysis widely applied in the electronics and automotive industries record a large number of parameters associated with the product. If the sum of the individual measured values exceeds the defined limit value, the product is prematurely marked as waste and the process may automatically stop. Thus, with the help of continuous process control, it is possible to avoid reworking and scrapping scraps and move towards production with zero scraps.


Maintenance management




Maintenance management included in MES has the task of planning individual repairs and maintenance and implementing relevant preventive measures. Well-organized maintenance management in MES contributes significantly to the achievement of quality objectives. Since it is not possible to completely avoid failures through preventive measures in automated production, it is necessary to have a sophisticated alarm system with a short response time to achieve high availability values of machines and equipment, which is also supported in MES.


Prescriptive maintenance and repairs


Prescriptive maintenance based on machine conditions


Current maintenance management (TPM Total Productive Maintenance) is increasingly going the way of prescriptive maintenance, where machine maintenance is deployed according to machine conditions. Here, online factors such as vibration, energy consumption, heat release, etc. are monitored. If the permitted limit values are exceeded, a maintenance order is released or a warning message is sent to the machine terminal.


Prescriptive maintenance based on the use of the machine or equipment


The machine or equipment usage time is recorded in the MES using the production function during order processing. Machine or equipment usage can also be recorded using, for example, the number of work cycles. If pre-defined usage thresholds are reached, such as the number of hours of use or the number of products on the piece counter, the MES can automatically generate a preventive maintenance order or display a message on the terminal.


Alarm management


If an unexpected shutdown occurs that cannot be resolved by the machine operator, the operator needs to generate a maintenance order for the maintenance and repair department within the MES. In fully automatic systems, this is secured directly using alarms. The responsible employees are notified via e-mail, SMS, or directly on the terminal screen. In order to monitor the availability of individual machines, their shutdowns are recorded with the justification of the operator on the terminal using a list of possible faults assigned to a specific situation.

Prescriptive maintenance of machines and equipment based on real-time data using the MES system analyzed using machine learning (Machine Learning).