Predictive analytics is one of the most fruitful disciplines in chemical engineering when talking about building your own business, whether a start-up or a consulting one.
I believe you have heard about that before, but maybe still have no idea about - HOW?
Let's first explore - WHY? After that, let's also explore - HOW?

Black box modeling usually refers to application of neural networks model. However, black box modeling also includes from simple linear models to partial least squares modeling, time series modeling, fuzzy modeling and neuro-fuzzy modeling. All of them have common approach discussed in this article. With plenty of simple tools available, you can start using this straightforward approach to help you solve complex industrial problems from today!
 

It is not the question anymore if predictive analytics can improve the plant operation - the only question left is "How much?"

These days everyone is talking about IIoT and big data, but do we really know what those are capable of doing when installed on real plants? Do we have a full picture of their practical benefits and potential improvements?

Process history and real time data hold the huge power if they are exploited in the right way. They are essential for optimal operation of any plant.

Predictive maintenance is often misunderstood and misused program. Most users define it as a means to prevent catastrophic failure of critical rotating machinery while its role is to be a part of an integrated, total plant performance management program. As such, it can provide the means to improve the production capacity, product quality, and overall effectiveness of our manufacturing and production plants.

Chemical plants are usually highly instrumented and have a large number of sensors that collect measured data for process control and monitoring. About two decades ago researchers began using the large amount of data to build predictive models, and these model were called soft sensors. Soft sensors are most often data-driven models based on data measured within the processing plants, providing real-time information necessary for effective product quality control.
 

The volume of information available has grown significantly over the last decade and has open a totally new area of using the data to help both science and operations. These days everybody is talking about “big data” concepts.

We have looked into some of the data mining tools available for free use and have come to a conclusion – there is really a lot available!!! 

The necessity to operate industrial units at minimum operating costs and still meet all requirements of product quality and tight environmental laws requires continuous process monitoring, optimization and efficient quality control. Nowadays soft sensors are used as the supplement to online instrument measurements for process monitoring and control improvement. Given that the industrial applications must be simple and easy to use but still reliable, the objective of this article is to illustrate the application of the soft sensors (data-driven sensors) on examples of commercial plants.

Application of mathematical models for the purpose of process monitoring is rising due to improved support of data historization and data handling. In most modern industries, process control systems are connected to systems that collect process data on a regular basis (e.g. 1 minute) and by using special data compression techniques, data can be stored for years and used for purposes of improved monitoring and control. Models developed for process monitoring enable faults recognitions, prediction of properties and improved quality and process control.