On July 14th and 16th 2015 we hosted a Manufacturing Execution System (MES) Essentials webinar.
The goal was to answer the common questions we receive about MES systems.
What is an MES system?
What is OEE?
How do I use OEE to impact my costs/profits?
What are other common Key Performance Indicators (KPIs)?
How does my team access the information the MES system provides?
One question asked in both sessions was: "How do you help someone decide what MES platform is right for them?"
The answer is almost as open ended as the number of choices on the market.
Here is our process to help a customer choose an MES solution:
1. Does the customer have a preference for one system over another? This can be based on seeing a system in place at another facility, marketing materials, or just personal preference.
In some cases this can make for a quick decision process.
2. Is there a software platform already in place on the control system side? For example, they are already using Inductive Automation, Rockwell Automation, Siemens, Wonderware, etc.
If so it usually makes sense to stick with the vendor they already have. There are already software support costs, internal knowledge, and a foundation for integration between the two systems.
The caveat here is if they are not happy with the existing system for some reason. In this case we would keep digging deeper.
3. Are there any hardware compatibility issues with a particular MES system and the PLCs on the plant floor?
In most cases communicating between any MES system and any PLC isn't impossible. Certain combinations of software and hardware may prove to be more costly. For example, if there is an older PLC requiring serial communications, an MES system with this option built into the software is a good option. Contrast this with a system that without serial support, requiring a hardware interface at the PLC level.
The other cost for communication hurdles is an OPC server to link software to hardware. In single vendor systems this is usually not required, but can be the easiest way to bridge the vendor gaps in some facilities.
4. The final question geared towards existing infrastructure is what business systems and ERP system is in use on the business side?
Most MES systems can communicate with most databases and ERP systems, with the potential for extra integration time in some cases.
These questions help define the costs to integrate a particular MES system into existing systems.
Assuming at this point the options haven't been narrowed down enough we get into more specific questions.
5. How much ownership of the system does the customer want to have long-term? If the customer wants to maintain and grow the system internally, their level of technical competence can dictate the right option.
MES systems move beyond ladder logic and HMI development into the world of databases, web services, and IT infrastructure. If the customer does not have internal resources to maintain these systems, the choice for a more user friendly system is easy.
If instead the customer is comfortable using the system integrator for ongoing support, this isn't as big of a factor.
6. Does the customer want a single system with a wide array functionality, or many smaller systems with more specialized functionality?
For example, it is possible to buy a system with OEE, SPC, Track and Trace, Reporting, ERP integrations, and other KPIs and tools built into one interface. Or they can start with OEE and Reporting tools, then buy a separate system with SPC, one for Track and Trace, and integrate everything as the system grows.
This choice is usally dictacted by who "owns" the system. A single system is usually better for the process operations team to manage, while the multiple system approach can be better for IT to manage. Smaller systems "touch" less of the process at a time, requiring less domain knowledge about any one area. All-encompassing systems interact with more of the process and other pieces of the system, making it easier for someone familiar with the process to manage.
At this point, we have usually narrowed the options down to 2 to 3 systems. This brings us to the final question:
7. Overall Cost.
The cost of the system is a huge factor in deciding which option to select. This includes the cost of the software, networking hardware, and servers required to run the system. It also includes the labor and time to integrate the system with the shop floor and the top floor.
This is where the system archictecture can influence the lifecycle of the project. Smaller systems can be implemented in phases, spreading the costs of both software/hardware and labor out over time. Larger systems integrations can be phased, with a higher software cost up front.
Implementations cost is where the discussion of ROI comes into play with the cost/benefit analysis also influencing the decision.
Another question we got was about difficulty between production staff and IT on MES projects.
"How do we overcome the organizational hurdles between production and IT to integrate business and production systems?"
This is a common problem. Production needs to have certain network access to enable database connections. IT wants to keep everything secure and not open ports. Some software installations need administrator rights and users don't have them. IT doesn't know the ins and outs of manufacturing software and can't foresee potential pitfalls.
We overcome these issues by acting as the middleman or facilitator for discussions between production and the rest of the organization.
We have experience on both sides of the table and understand how to frame information and requests so the other side can hear them.
IT is not against making it easy for production systems to access business systems. They are against opening up the network to unnecessary security risk. They will give permissions to set up software when necessary, and will reduce them upon completion to ensure system integrity.
Framing the discussion in specific terms is the first step to get IT to buy into an MES project. The main questions you will need to answer are:
What access is necessary, physical, network, and user permissions?
What data is required?
What is what is the least risky way to access the data?
Putting together this information before sitting down with IT will ease frustration across the board.
Other things we like to do are to bring IT in once you have decided to move forward with an MES project. Keep them in the loop as you make vendor decisions. Introduce your integration partners to IT and start the relationship as early as possible.
The other hurdle to overcome is understanding system ownership. If IT will manage licensing, support, and configuration after the project is complete, involve them in the project. Include them in training sessions, and help them learn the production side of the MES system. The same holds true on the production side if production will take ownership.
The final question we would like to cover focused on the long-term use of MES in an organization.
"How do you ensure a company participates in continuous improvement over the life of an MES system?"
The foundation of continuous improvement is cultural within an organization. It requires business objectives to be tied to continuous improvement. It also requires internal champions to monitor, maintain, and even retire KPIs, reports, and other informational tools over time.
For example, a company is tracking OEE, and finds their main downtime issue is changeovers on a particular machine. Seeking to narrow down the cause, the add a KPI for changeover durations on this machine. After measuring for a couple of weeks they find the problem stems from a lack of training. After revamping the training class and re-training the operators, the issue is largely resolved.
In some companies, once this measurement has made it into reports, it will be in the reports forever. The thinking here is something was important enough to measure once should be measured forever. This type of thinking will eventually lead to information overload.
The goal of continuous improvement is to always improve your results. As your results improve, you will also need to improve your tools and processes to stay up to date with the rest of your process. This means it is important to use measurements for only as long as they are useful. Once issues are fixed and the measurements are no longer changing, they can be taken out of regular circulation.
Many times we have gone into a facility and they have dozens of pages of reports, and everyone only looks at one or two pages. These pages focus on the current initiatives and the rest have just never been taken out of the reports.
Continuous improvement is all about changing the idea of "that's how we have always done things."
We hope you enjoyed our webinar and were able to learn something.
If you have any questions, please ask us anytime, we are always happy to help.