On Artificial Intelligence and Ethics
The Tango of AI and Big Data
Transforming the Art Museum in the 21st Century.
Embedding AI in everything. Why we can't do without it now
When it Comes to AI, the Fun Part is Over
Beaumont Vance, Director of the Analytics Center of Excellence, TD Ameritrade
Applications of Big Data and Predictive Modeling
Somesh Nigam, Ph.D., SVP and Chief Data Officer, Blue Cross and Blue Shield of Louisiana
Grow and Supercharge with Artificial Intelligence
Deep Varma, VP of Engineering, Trulia
Technology Megatrends that Will Shape the Future and Your Business
Jag Randhawa, Technology Executive & Author, CAMICO
Power of Machine Learning in Plastic Processing
The plastic injection molding process has a number of variables such as pressure, injection velocity, barrel temperature etc. In order to have a high-quality product it is crucial to keep the whole set of factors at an optimum level. The optimization through conventional mathematical modeling is an extremely challenging process because the lack of consistent relationship between the process parameters. Thus the traditional approach uses manually prefigured configuration based on rigid threshold levels. But today an innovation exploiting machine learning algorithms provide a new regulation tool which allows optimizing the whole set of variables in real-time.
Machine learning doesn't rely on domain expertise unlike conventional approach adjusting parameters based on the operator's experience. The main source of knowledge is the historical data recorded by sensors from the particular molding equipment. After processing such sets of data machine learning learns complex relationships between parameters and quality of the product. That is, the system builds the required prediction mathematical model by its own.
Training of the machine learning model can predict the final quality of the plastic part by processing live data from sensors. Factors such as quality and volume of training data, level data preparation and cleansing, chosen machine learning algorithms, and the experience of data scientists determine the accuracy of the prediction. In some cases, machine learning models are capable of distinguishing the predict types of faults, for example, unfilled areas, warped parts.
This new insight can be used in two different ways:
1. If a machine learning predicts poor final quality, the manufacturer can stop the further processing.
2. Machine learning models provide an operator with additional precise insight into the process allowing further optimization of the molding parameter optimization.
Another implementation of machine learning technology is the predictive maintenance system. Downtime reduction is a perk of predictive maintenance which can prevent decreased production and wastage of raw materials. The predictive maintenance also raises an early warning for critical equipment failure.
Business leaders in the plastic industry are realizing the potentials of machine learning to reduce costs, and improve the overall quality of production.