Maintenance software that has real-time machine and operating data at its disposal improves overall equipment effectiveness (OEE), minimizes downtime and reduces the workload on maintenance staff. Digital maintenance also offers the following benefits:
1. Improved Economic Efficiency
- reduction of downtimes
- reduction of costs for unplanned downtime
- increasing the service life of machines and systems through regular demand-oriented planned maintenance
- better predictions for spare parts management
- for the machine through permanent reporting of machine data
- within the production process through planning
- increased flexibility in production
- through permanent analysis of the collected data
- in the long term: achieving higher productivity
- faster fault detection and response to malfunctions
Downtime is the enemy of profit. Disruptions in the production process or even system failures in production have a direct impact on a company’s operating result. Digitalized maintenance processes and optimized maintenance intervals ensure more efficient production processes.
Steffen Dams, Team Leader Consulting at FASTEC
A Combination of Several Approaches Leads to the Goal
Basically, there is event-based, interval-based and condition-based maintenance. The goal is often only achieved through a combination of these. With event-oriented maintenance, a response is made when a failure has already occurred. This can be an acceptable solution if the failure of the system does not result in high costs due to consequential damage or loss of production and if the application on the machine plays a subordinate role in the production process. The procurement time for spare parts is also a decisive factor here. If these are available at short notice, reactive maintenance can also be implemented efficiently.
If the downtime costs are much higher than the costs of planned maintenance, the main focus should be on preventive or interval-oriented maintenance. With preventive maintenance, critical wear parts are replaced at fixed intervals. Predictable downtimes can thus be avoided. Planned maintenance is based on the average expected service life of individual parts or systems in order to predict when maintenance will be required. Planned maintenance is carried out at predetermined (regular) intervals, for example every 8000 units or every four weeks. The maintenance measure is therefore based on a theoretical failure rate.
However, the actual system performance, i.e. the actual wear and tear, is not taken into account. The actual wear and tear on a system differs greatly in terms of its actual use, so that preventive maintenance can result in unnecessary additional maintenance work. Or parts are replaced early, in line with the statistical maintenance interval, and then scrapped, even though they are still working perfectly. This is cost-intensive and is not in line with the idea of sustainability and resource conservation.
In predictive maintenance, process and machine data is reported in such a way that forecasts become possible. Systems can be maintained proactively or in line with demand before a malfunction occurs so that downtimes do not occur in the first place.
Steffen Dams, Team Leader Consulting at FASTEC
With predictive maintenance or condition-based maintenance, machine parts should be used for as long as possible and only replaced shortly before their foreseeable failure. For example, individual areas of the system at production-critical points are monitored digitally via sensors during operation. The sensors monitor the machine’s condition by measuring temperature, vibration, humidity and pressure. Deviations in vibration, for example, are an indicator that a machine part is about to wear out. The machines are linked to each other, enabling the timely identification of error sources and prompt intervention. Faults can thus be diagnosed with a high degree of accuracy and planned maintenance and machine downtimes can be planned precisely.
The main aim of predictive maintenance is to plan maintenance as precisely as possible in advance in order to avoid unexpected system failures and therefore unnecessary costs. Knowing when which machines or individual parts need to be serviced means that resources for maintenance work such as spare parts or working hours can be better planned. Plant availability can also be increased by converting “unplanned stops” into ever shorter and more frequent “planned stops”. Additional opportunities include a potentially longer machine service life, increased plant safety, a reduction in accidents with a negative impact on employees or the environment and optimized spare parts handling.
Our Maintenance module in the MES FASTEC 4 PRO provides practical support for maintenance. Functions such as the creation of autonomous maintenance plans or the documentation of activities performed and spare parts required ensure, among other things, significant time and cost savings and reduce the workload of maintenance staff.
Steffen Dams, Team Leader Consulting at FASTEC
Data Quality is Crucial for Predictive Maintenance
With continuous data acquisition, data evaluation and complete networking of systems, companies in the manufacturing industry are already very close to their goal of near-zero downtime production. However, whether predictive maintenance works depends heavily on the available data quality: with an MES and modules for Machine Data Acquisition (MDA) or Production Data Acquisition (PDA), companies can collect and evaluate relevant data and automate maintenance intervals. If the maintenance software is part of an MES software solution, they have access to a large amount of qualitative machine data – in the form of real-time data as well as historical data, as this is automatically documented digitally. Networking allows condition data, consumption and wear values, for example, to be recorded, providing a very accurate image of the state of the systems and the parts produced.
When effectively combined with data from the MES and ERP system, maintenance times for a system can be precisely predicted. The software continuously controls the use of machines and tools and automatically reminds the maintenance technician when planned maintenance is due. Rigidly scheduled, regular maintenance intervals are interrupted and the maintenance technician can intervene before the productivity of the machines threatens to decrease – or only when it is really necessary.
A shared database offers the following advantages:
- Existing resources such as personnel groups, equipment and work centers/lines can be used in the entire production environment without interfaces or duplicate data storage.
- The machine operator selects a predefined state. This automatically creates a maintenance order and sends alerts to the people/departments involved.
- When the maintenance technician starts planned maintenance, the appropriate machine status is automatically triggered and recorded by the MDA.
Conclusion & Recommendations for a Successful Implementation
In order for companies to make reliable statements about the state of machines and systems and thus identify expected faults, they need to collect a large quantity of data with an MES such as FASTEC 4 PRO over a longer period of time. Data analysis is used to analyze past scenarios in order to identify sources of error and implement future improvements.
In combination with artificial intelligence, predictive analytics makes it possible to avoid faults long before they occur. Based on historical data, it is possible to make predictions about future events that can be traced back to certain constellations of parameters (e.g. machine settings in production). A solid foundation of digitally recorded production data (MDA, PDA, Maintenance) is therefore a prerequisite for future AI projects. The larger the database and the more intelligent and sophisticated the algorithm, the more reliable the findings will be.
These steps will help you to successfully implement digital maintenance:
- Start with small, focused projects for a critical system or a critical process
- Define data elements and technical requirements for the use case
- Collect relevant data
- Iterative optimizations
- Gradually scale selected systems
- Retrofit relevant systems to collect additional data and implement further optimizations
- Build up expertise (internal, external)
- Consider IT security to prevent unauthorized access
