Our Kickoff with Industry 4.0. A start towards live data control with Machine Learning in the Industry. How to control and improve a process on the fly? We will have a Live Demo, including multiple applications working together to sustain a solution. Our team will share with you the difficulties and techniques that were used to achieve this.
PROGRAM AT CONET
18:00 – Food & Beverage
18:30 – Introduction
18:40 – Talk 1 ”Applying Data Science for live Responsive Actions.”
19:30 – Short Break
19:45 – Talk 2 “Connected Factories, from the factory floor to the cloud”
20:30 – Drinks
TALK 1 “Applying Data Science for live Responsive Actions”
Speaker: Laurens Koppenol
The food and beverage industry copes with huge but unpredictable demand. Keeping the balance between meeting production and quality requirements is a constant struggle. Typically, operational processes are complex and require in-depth knowledge, i.e., years of experience. A solution was developed using the synergy of domain experts and data scientists, reducing food waste and increasing efficiency.
In the production of lemonade, quality is measured after a batch has finished. A batch has to correct or disposed if certain quality criteria are not met. Using machine learning to continuously predict the quality and change the process accordingly, seems like a straight-forward improvement. This does come with additional challenges, which are representative of most industrial manufacturing processes.
Coming to this talk, you will learn how to turn your manufacturing challenge into a machine-learning ready optimization statement. You will also learn how machine learning can help solve this problem and how this can be applied in the real world.
You will be able to see how a model is trained and applied to a simulated environment. If you are lucky, you might even be able to taste a perfect, freshly produced, lemonade.
The presented solution uses a combination of the following technologies: Siemens PCS7, Python, CatBoost, Docker, Azure Machine Learning Services, Azure IoT Hub.
TALK 2 “Connected Factories, from the factory floor to the cloud”
Speaker: Keagan Ladds
Industrial process control can produce a large amount of data, often providing critical insight into the short- and long-term operation of a plant. However, there is typically a distinct divide between the factory floor and the platforms used to make sense of all this data. This divide provides a challenge in extracting the data quickly enough for it to provide actionable insights and is often only used to provide an after-the-fact analysis of the running of a plant. The use of an IoT platform that enables the deployment of a collection of containerized applications to the edge helps to bridge this divide.
By attending this talk, you will learn how to make use of edge deployed containers to provide on-premise processing of factory data, as well as how to stream actionable insights to the cloud in near real-time. You will be able to see how containerized applications are deployed to the edge to stream process information from a PLCS7 system to the cloud. You will also be able to see a dashboard, complete with changing gauges, numbers, and graphs.
The presented solution uses the combination of the following technologies: .Net Core, Siemens PCS7, OPC UA, Docker, and Azure IoT Hub.