At the cyber – FoF’s core very essences are Reliability and Intelligent software. Without them:
- Operation performance is unstable.
- Operations technology fundamental laws are ignored.
Consequence: With a high automated – sophisticate environment, a very poor plant quality and throughput.
Reliability in modern conception means near 100% confidence on:
- Product quality and life–span
- Process repeatability, and
- Machine continuous repeatable running.
But with an holistic conception among them. I mean, that a permeable transmission of clever – filtered technological info preventing yet problems and improving future, must be set up between product, process, and maintenance.
The role of Promind®, the Prisma®’s A.I. module, is to be this intelligent filter, with two kinds of:
- Tactical: real-time predictive alarms, their causes, and suggested solutions
- Strategical: background self-learning knowledge over the Math-Physical laws behind product-process and machines behavior.
Both on high-easy interaction with engineers for creating-evolving models and assessing the provided info in terms of practical results.
Promind® with its two essential modules Sys CL + PML, displays the innovative technique of Symbolic Regression for better modelling in a comprehensible way the involved underlaying physical laws along the treated problem.
All the involved information is related with well-know Manufacturing Tools, in order to be easily manageable by the engineers.
Some of them, as example are:
- Weibull distribution and inference maths.
- Wöhler curve for fatigue testing
- Failure trees, cause-root analysis
- SPC and quality management tools
- Euler-Lagrange dynamic optimization models.
Any new clean energy or greener fuel generation plant need to graze perfection. As a commodity factory dealing with a very sensitive subject, must perform value creation evidence along operation. From solar pads to hydrogen or new advanced nuclear power plants, the commons are:
- Sophisticated equipment
- Full safe automation with adaptive-predictive PIDs
- Continuous running round the clock
- Capital intensive
- Need for an extreme performance and efficiency
- Need for extreme reliability
Performance and efficiency are measured by 4 basic inter-dependent elements that must be controlled in real-time:
So we need fix items by design, otherwise no operational practice could improve, and have done it apply very special managerial habits, appropriated for a context in which there’s a very little margin for error and delays.
Note that the loop Q – OEE – C is self dynamic – contained, requiring intelligent precise automation to be real-time balanced, and that overall waste reduction, a symptom of a fine operation procedure, is an “external” force that can improve or deteriorate such balance.
The design conditions that help to make the loop Q – OEE – C stable at a higher throughput level are the following:
- Extreme high quality equipment items
- Robust, rigorous and simple integration design: question each nut and screw!!
- Design for reliability of the loop Q – OEE – C
- Real-time predicitive reliability of Q – OEE – C , this needing the use of A.I. reliability focused software such as Sisteplant’s Promind® and Prisma® (CAMT)
- Process + Maintenance Engineering as an holistic unique block with a given methodology such as TIM (Tech Integrated Maintenance) of Sisteplant
Note again that the last two need a Scientifical Maintenance as the only feasible path of the required holistic integration.
What does it mean Scientifical Maintenance?
- Deep process and equipment physical laws dominion
- Math and A.I. models design and results a knowledge
- Predictive managerial as a daily practice
- Practical approach linking failure in any Q – OEE – C – W with those physical laws, and so working out much more sounded solutions and adjusting the models.
Where and how can Sisteplant help to you in the configuration and exploitation of an advanced facility like these?
- In assessing the coherence and rigorous completion of the Q – OEE – C balanced loop design
- In the definition and implementation of the Scientifical Maintenance Required Model
- In the practical implementation of the intelligent simulation and Maintenance technology management with Promind® and Prisma®
- With the MES software Captor® that controls and links with Prisma and the process PIDs regulators all the plant events not directly related (though relevant) to the process variables evolution
With a commitment with the setting up of an effective technology driven organizational model much more beyond the mere classical ICT implementation.
It is an strategic sofware piece for the cyber factory. Traditional MES are more operative systems filling the gap between the MRP-2 and the real-time that imperates in the shop floor, but the fact of having in Captor embebbed-friendly intelligent algorithms convert it in something valuable difference. Here’s why.
The sense of the cyber factory is not only advanced flexible automation and its main cast the robotics, but the intention of having permanently under real-time control by humans its intrinsinc natural concurrent events as well.
These events range from horrizontal and vertical SPC warnings and predictive machine and process reliability to flexible flow optimization and lead-times alarms.
The think is not to pretend a model in wich the computer decides what to do using the A.I. and traditional advanced algorithms that Captor manages. This, through complex enough to the programmed, is not useful by itself, because would represent only tactical actions.
What means that? Single tactical actions lack the intelligence of strategical high-level mind decissions only qualified humans can have. So, the challenge that Captor® – Prisma® has solved is to give people the filtered capacity to concurrently interact with the low optimized level decissions presented by it’s A.I. models, calculus that humans can’t compute with precission nor the agility of the real time, as well as the criss-cross complications of integral alarms over process, maintenance and flows.
The advantage of that is confidence and acceptance by engineers and plant operators, as well as (very crucial) the capacity of the Captor®-Prisma® models to train people in the roots of process and reliability variations.
So the expectations of an I-MES like Captor® over other plane MES systems is that in a correct implementation:
- Quality improves directly at least 50% more.
- Machine breackdown is directly reduced al least by a 80%.
- Lead times are reduced by at least a 50%.
- Engineers and operators grow in their technical level, putting them in a plane un-reachable by robots.
The implementation steps are so more delicated and important that with traditional MES, but it worths because all of that means a direct pay-back of 1 year or less.
CONOCE LOS COSTES REALES DE FABRICACIÓN Y HABILITA UN CUADRO DE MANDO REALISTA QUE APOYE LA TOMA DE DECISIONES ÁGIL
En toda situación, y particularmente en tiempos de incertidumbre, es clave disponer de un cuadro de mando realista que apoye la toma de decisiones.
Is digitalization suitable for SMEs? Any company of any size can define a deep industry 4.0 transformation roadmap, or this is just a big players game? How to find what it takes to start a rollout?
Small and medium-sized enterprises (SME) are at the core of business development in Europe. 25 million companies, producing 50% of European GDP and 2 out of 3 jobs are coming from SMEs. Digitalization should be the cornerstone for business success but somehow we need a final push to jump from the springboard, when it comes to implementation.
What it brings. Clarifying goals to be achieved.
Opportunities coming from an Industry 4.0 roadmap are on everyone’s lips, but sometimes it’s not so easy to find arguments to justify our own case and move ahead. Here are some clues that can be helpful in order to understand what digitalization brings to a SME.
|Cost reduction||It’s easy to think that one of the first source of benefit that can be identified is close to direct labor cost reduction. Automatization of manufacturing operations and tasks performed by workers comes to direct headcount savings.
Nevertheless, the impact in indirect labor is also relevant. Information gathered from the shop floor is downloaded into a data lake creating a structured plant information map. Dashboards and both real time and historian analysis tools are deployed to help decision taking. This deployment comes with the reduction of non value tasks and the optimization of indirect labor.
|Increase of plant capacity||OEE and LTA (lost time analysis) metrics allow to identify hidden sources of waste.
Plant teams are trained in lean manufacturing techniques and waste reduction initiatives become part of day to day factory culture.
New machine capacity arises as waste is been reduced.
|Resources usage (water, coolants, energy ….) can be easily gathered from IoT devices. When included in the equation, sustainability metrics are calculated and so the impact on the environment can be tracked.
Very often environmental care can easily be translated to cost saving figures.
|Customer loyalty||Production follow up, idle times identification and wip optimization come to lead time reduction. Depending on the manufacturing strategy (product structure and process flow) savings can be huge.
Variance reduction (with 6sgima, SPC or deep knowledge models) increase process robustness and therefore lead to higher quality standards.
Both lead time and non-quality reduction are key factors to increase customer satisfaction.
People development and organization growth
|Data becomes a new raw material. The smart use of factory data leads to new disruptive opportunities.
o Machine learning models based on real data allow to create explicit knowledge on how processes work, and to deploy this knowledge organization wide.
o New capabilities are developed in existing workers mixing up both manufacturing and IT skills.
o Business as usual is continuously questioned and innovation processes are part of day to day job.
o Plant is a technological entity plenty of attractive for new talent recruiting.
Adding the specifics of each industrial sector and the maturity level of each company, the above arguments, and for sure many others, can be the basis for setting up a reliable transformation project.
What it takes. Best practices tips
Transformation roadmap is not an easy process and there are many decisions to be taken that can make the project fail. There are some tips that can help in the early stages
- Make your customer the hearth of the process. Market trends and customer needs are changing. Define your goal starting from your market/customer requirements and then design the transformation roadmap
- Think Big. It needs to be challenging. No limits to creativity must be set. Opportunities can be hidden anywhere.
- Think global. Transformation process demands and holistic approach to lead to success. Think global even though you have to start local.
- World class transformation. Digital transformation needs to be combined with and organizational transformation based on world class manufacturing principles and methods.
- Full commitment. Involve the whole organization from manager board to workers
- People Centric. Technology is not the goal is just an enabler. The competitive key is how our teams are taking advantage of new techs deployed. People transformation is mandatory since Human Factory is one of the key axis.
- Set up a partners network. Industry 4.0 enabling technologies are evolving really fast. A wide collaborative network will allow accessing digital specialists when needed.
Once our goal with digitalization is fully understood a tangible case study can be set and will clarify the benefits of deploying a transformation roadmap. Financial figures based on short term benefits are needed, but beyond this, there are insights that if properly focused to cultural change that can boost ROI. Don´t forget to add them to the magic formula.