A modern plant is highly complex production machine, ruled by predictable causes and time-delayed effects. But like any truly complex system, full grasp of everything that happens in it defies the ability of any one human. Knowing where and what the optimal running state of the complex system is at any moment, 24 hours a day, 7 days a week, is beyond even the most skilled team. It is no wonder then that sub-optimal operating procedures, poor design or improper scheduling of production have been identified as the main sources of many plant reliability problems. But nowadays, we measure everything that happens in a production cycle – every action, every procedure, every step of the process, is observed by sensors. Then all this big data, gigabytes per day, is digitally recorded on hard drives and largely forgotten about. We exist in a reality of big data, but little insight.
The power of the Advanced Process Optimizer (APO) is in its ability to use advanced machine learning algorithms to model and optimize your plant not based on physical principles, but solely on the data. We translate your plant, your problem and your goal into mathematics, input all the data, and solve it. Using this approach, we don’t need to understand every process in your plant, or model the state of every machine – that information is already in the measurements. And while the model is in itself highly complex and technical, you experience none of the complexity. Once APO is installed in your plant, it delivers simple, actionable suggestions to optimize the running of your plant. By doing this, APO can increase the output of your plant by up to 5% and will pay for itself in less than a year.
Even better, APO is not a static model. Thanks to its machine learning, once implemented in your plant it observes and continues to learn in real-time. As the conditions in your plant change, so does APO’s understanding of it – and without any need for you to update it.