Given how fast technology and markets are changing, how can manufacturers of intelligent products ensure they are ready for what’s next? Future-ready manufacturing requires plant floor systems that go beyond traditional manufacturing execution systems (MES). With the full scope of manufacturing operations management (MOM), analytics, and artificial intelligence (AI), companies have a good opportunity to be agile enough to make the next pivot. This eBook explores all of that and more.
Please enjoy the summary* below. For the full research, please visit our sponsor iTAC Software (registration required).
Table of Contents
- MES is Evolving
- MES, Analytics, and AI for the Journey
- End-to-End Operations Coverage and Control
- Functional Breath
- Connectivity for Data Flows
- Connector Characteristics
- Modern Future-Ready Architecture
- Analytics to Extend Data Value
- Flexible and Agile to Match Operations
- Suitable for Multi-Plant Collaboration
- Beyond Today’s Realities
- Future-Ready MES Considerations
- Acknowledgments
Foundations for Future Success
Comprehensive Operational Software
Given how fast high-tech products, materials, suppliers, and processes change, how can the manufacturing operation keep up and achieve business goals? Being future-ready is complex and multi-faceted, particularly for serialized high-tech manufacturing. These companies have particular requirements for manufacturing execution systems (MES) and the expanded functionality called manufacturing operations software (MOM).
Ensuring the MES will evolve into the future has additional requirements. One of those is leveraging the data more fully for analysis. Combining MES with analytics and artificial intelligence (AI) can significantly enhance each technology’s current and future value.
MES Is Evolving
MES Expands to MOM
The face of modern MES is different than it was even a few years ago. One element is that the scope is more extensive. This is often considered an expansion from MES to MOM, with functions to support not only production but also quality, maintenance, and inventory inherently in the system, per the ISA95 model.
More Data
Makers of smart products typically also have far more data to process. MES must be ready to handle data feeds from equipment and industrial Internet of Things (IIoT) devices. In addition, many products are more complex than ever, with more variants, and all of these have serialized components at every level.
MES for Quick Value
Manufacturers want to gain value from software immediately. To accomplish that, companies are moving to more tailorable applications. With tailoring, software providers can focus more on the product than customizing code and laboring through each project. This means manufacturers get a good fit and quick time to value.
Ready for Change
Software must be flexible and adaptable to keep up with the pace of change in manufacturers’ products, materials, processes, and people.
Beyond the physical, data requirements are also changing frequently. These changes come from shifting expectations for customer data, cybersecurity, sustainability, and new regulations.
Analytics for Better Outcomes
MES has always been a source of critical in-context data about operations, products, and processes.
Performance dashboards and current-state views have long been part of MES. Today’s analytics and AI can sit inside MES to drive better operations and business outcomes. MES with advanced analytics and AI can also help troubleshoot causes, predict what will happen, compare options, and guide more effective action.
Analytics to Extend Data Value
Analytics in MES
Aim for data-driven decisions from the system for operators, data scientists, and engineers. MES with analytics pre-configured functionality is common. Beyond that, look for an open analytics platform for development. This enables third-party tools to come into play and supports multiple parties collaborating.
AI and ML
Modern software should include an engine for machine learning and artificial intelligence (ML/AI). Look for models trained based on the MES provider’s expertise. The AI capability must span specific production environments that the engineers know, not just data scientists. The beauty of ML, as the name suggests, is that it learns from reality. In this case, the MES data reflects reality and teaches the model. Sometimes, the MOM offering also includes a toolbox of algorithms that can deliver automated insights.
At the Speed of Operations
What’s crucial is that the analytics, AI, or ML provide results fast enough for the operation to use them effectively. Event streaming can add speed and agility. For maximum effectiveness, the system will also deliver alarms and escalation when anomalies occur. Decision-makers often need to know right away.
Optimizing Analytics
While basic analytics are standard for MES, some special factors can optimize analytics. These include:
- Data quality and model or algorithm validation
- Analytics that process engineers can configure, not just IT
- Digital twin for testing algorithms and approaches
- AI support for system users
Beyond Today’s Realities
People-friendly
MES does connect to equipment, but human interaction is crucial. We need humans in the loop for anomalies and new situations to ensure AI is on track. This means the system should provide visualization of data and status through intuitive UIs. Ongoing changes in reality mean the system must also be simple to set up, use, and change.
Analytics and AI
We are already in the age of analytics and AI in production operations. However, moving beyond where we are today will require a robust approach to analytics in all forms. For example, prescriptive analytics, which recommends actions to take in the future based on the past, will benefit from complete data sets and history. More sophisticated ML pattern recognition will likely also come online to spot anomalies in real-time and trigger alerts and action to prevent scrap and rework.
Autonomous Future
Many high-volume production tasks are tedious and repetitive. Some are automated, but many plants are not automated beyond a particular line, such as an SMT line. Autonomous operation for repetitive tasks and workflows is likely in the future. This will require an event-driven architecture that connects to every piece of equipment, data source, and data-consuming device in the operation. It will also require flexibility to change as conditions change.
Ecosystem Connection
There is also hope for industry-wide data connectivity standardization in the future. Organizations such as International Manufacturing-X6 are building out use cases for using industrial data across supply chains. The need for multi-company analytics and traceability is crucial, and software should consider new guidelines as it evolves.
*This summary is an abbreviated version of the ebook and does not contain the full content. For the full report, please visit our sponsor iTAC Software.
If you have difficulty obtaining a copy of the research, please contact us.
