How we optimize Manufacturing Process Intelligence
Andrew Sparrow
Driving Supply Chain Excellence: Integrating Advanced Manufacturing, Data Analytics, & Sustainability Initiatives for Resilience & Agility. Consultant | Speaker | Author | Live Shows. The Product Lifecycle Enthusiast
With huge amounts of data being extracted from the Shop Floor and the sheer computing power to analyze and optimize, many want to move their MES/MOM data to the Cloud.
An MES is a large, divisible system. It’s made up of sub-components that interact with each other and with the outside world to perform a service. Which of those components however can you afford to lose if your internet connection fails? Some? All?
Parts of the MES could live in the cloud and likely survive a reasonable amount of interruption. If you can’t respond to a quality incident and discover the identity of an imperfect intermediate batch right away, you’ll probably survive (although this will change as your systems integrate closer to those of your customers).
If, however you're unable to download process orders from your ERP and execute them, you’ll have a serious problem unless you have a means of creating and starting orders locally.
Then imagine, as we all move to advanced control of plant operations that's adjusting process sequencing, across grades in real-time and loss of internet in the middle of that process - yes, it becomes hugely disruptive along with damaging to the bottom-line.
So, What’s the right way to balance the needs of the plant floor with the advantages of cloud technologies?
Welcome to the hybrid on-premise/cloud approach, leveraging the best of both worlds.
Hybrid on-prem/cloud MES
With a hybrid on-premise/cloud Manufacturing Execution System (MES), your operations work on premise, and your analytics and optimization tools are applied only to the required subset of the data in the cloud. This is a cost-effective MES infrastructure as it does not pose a risk to production yet utilizes the cloud to help reduce the hardware CapEx.
Widespread adoption of hybrid on-prem/cloud MES will help manufacturers scale operations globally with complete operational visibility and control.
Data, data everywhere
According to the World Economic Forum, it is estimated that by 2025, 463 exabytes (1 exabyte = 1,000,000 TB) of consumer data will be generated every day globally. Factories are expected to add 22 Zettabytes (1 zettabyte = 1000 exabyte) of data on top of this.
The challenge is to make use of the data in a useful way and use it to deliver value to the business. However, today, a significant portion of this data is in the shadows as most environments are not meant to handle the volume and variety of data generated. Here comes the Cloud
The move to the Cloud
MES migration to the cloud is following a top-down approach along the ISA95 Automation Pyramid. ERP and PLM systems were the first software to migrate to the cloud, and now MES/MOM systems are increasingly moving there.
A hybrid on-premise/cloud architecture enables manufacturers to flexibly run control, computing, analysis and visualization tasks where most appropriate. In the context of MES, that means that production execution can be effectively run at the edge while longer term storage and data analytics heavy-lifting is done in the cloud.
There are two very different ways in which these software suites are migrating to the cloud:
- Lift & Shift (hosted private): The existing software is “lifted” from on-premises and hosted on a private or public cloud infrastructure located off-premises.
- SaaS (public): The software is “cloud-native”, meaning it is multi-tenant and built on the cloud managed by the software vendor.
The choice of the direction to go, comes down to the reliability of your internet connections at an operational level and if you're prepared to risk operations. Typically we're finding most customers prefer the blended On-premise MES solution with Cloud-based "Lift & Shift" of Manufacturing Process Intelligence.
Justify the Hybrid On-prem/Cloud MES
Moving from Preventative to Predictive to AI driven Prescriptive Maintenance
A hybrid on-premise / cloud MES solution supercharges analytics capabilities by integrating data from multiple systems, creating an enterprise data set for reporting and analytics. Companies can perform analytics on a richer, contextualized dataset that leverages industry-standard data models. The hybrid on-premise / cloud MES can move the data collected (both raw and contextual) to data lakes to pool data into a single location, making it easy and fast to create a context for manufacturing analytics and improve operations.
Just faster all round!
Each day, factories run faster by storing operational data in the cloud for analysis instead of storing on site. Your operators aren’t held up by MES systems struggling to cope with a large volume of on-premise data for analysis. The entire factory just runs faster.
Scalable intelligence
Manufacturers also unlock new ways to combine and view data remotely, comparing Manufacturing Process Intelligence dashboards across multiple plants, and tracking from the enterprise level to the shop floor. This helps every team make the best decisions based on the best data – faster.
Today’s MES technology harnesses invisible data and makes it easy to give each person visibility to the information they need to do their job.
Last, but by no means least
Yes, it saves you money!
Storing years of operational data is often a compliance requirement, but on-site server costs quickly add up. It is now much cheaper and affordable for manufacturers to keep a required subset of their OT information in the cloud – reducing the need for on-premise server storage.
Smarter Innovation & Product Lifecycle Management & Manufacturing: People, Teams & Business Solutions enabled through Change & Technology
Sometimes you need a real expert to help decide what's next and sometimes you need an entire team and sometimes you need an entire program delivering.
Delivering the entire PLM & Smart Manufacturing application layer, along with integration to ERP and moving your people to adopt new ways of working, is the holistic approach we take. It's the quality of our people and their experience that makes the difference.
If we can help you through your PLM & Smarter Manufacturing journey, you just have to ask
I'm a huge believer in constant change.
Standing still is going backwards
It starts with People changing their mindsets & Processes, enabled through Technology.
Smarter Innovation & Product Lifecycle Management & Manufacturing: People, Teams & Business Solutions enabled through Change & Technology
Sometimes you need a real expert to help decide what's next and sometimes you need an entire team and sometimes you need an entire program delivering.
It's the quality of our people and their experience that makes the difference.
Senior Manager of Program Management Departments, Contract based Program & Engagement Manager and Principal Professional Services Strategy Advisor (Medycus/6by6,llc)
3 年Andrew, thanks for another interesting article. With all of the recent hackings and ransomware issues on both cloud and on premises solutions and systems, as well as all of the distributed denial of service attacks, how does this change the optimal data and systems architectures for a hybrid cloud solution? Data needs to be encrypted at rest, encrypted during secure transmission and reside in an encrypted state at the receiving end. Even Windows Kubernetes Container files are vulnerable. Environments, whether they are in a private cloud or on premises, need to be secure, and properly monitored and properly operationally supported, and assessing a cloud vendor’s ability to perform these last two are difficult, even if the vendor is compliant to NIST or ISO standards. All of the above do you no good if the physical layer internet connectivity to the plant goes down in a hurricane, etc. Outages of a factory line could cost some large companies as much as $10 million every 15 minutes the system is down. Some sort of failover is clearly indicated that allows orders, data and instructions to flow on premises and the lines to continue manufacturing. This data will have to be captured and resynched once service is restored.