Industry 4.0: Machine Learning in Manufacturing

Which of the following is a supervised learning problem?

• Grouping different operating modes of a machine

• Predicting component failure based on historical data

• Predicting quality deviation based on historical data

• All the above

Robotic tasks include a multitude of ML methods tailored towards navigation, robotic control and several other tasks. Robotic control includes controlling the actuators available to the robotic system. An example of this is control of a robotic arm in paint-shops in automotive industries. The robotic arm must be able to paint every corner in the automotive parts while minimizing the quantity of paint wasted in the process. Which of the following learning paradigms would you select for training such a robotic arm?

• Supervised learning

• Unsupervised learning

• Combination of supervised and unsupervised learning

• Reinforcement learning

Suppose you are a quality engineer at a shop-floor. You have been given a task of making a system that recommends quality of components based on activities of a machining process. You realize that materials being used, and machine-tools parameters could be highly variable. Hence, you would decide to:

I. First, cluster machine parameters into operating modes and

II. Second, train separate models for each operating mode to predict quality of products.

The first task is a/an ______________ learning problem while the second is a/an ________________ learning problem. Choose from the options:

• Supervised and unsupervised

• Unsupervised and supervised

• Supervised and supervised

• Unsupervised and unsupervised

Which ONE of the following are regression tasks

• Predict the age of a machine

• Predict the OEM from where the machine has been purchased

• Predict whether a component of the machine will fail in the next 2 weeks

• Predict whether an operating mode relates to a specific machine

In k-NN algorithm, given a set of training examples and the value of k < size of training set (n), the algorithm predicts the class of a test example to be the

• Most frequent class among the classes of k closest training examples.

• Least frequent class among the classes of k closest training examples.

• Class of the closest point.

• Most frequent class among the classes of the k farthest training examples.

Which of the following is true for a decision tree?

• Decision tree is an example of linear classifier.

• The entropy of a node typically decreases as we go down a decision tree.

• Entropy is a measure of purity

• An attribute with lower mutual information should be preferred to other attributes.

Predictive maintenance involves:

• performing maintenance activities in a consistent, predictable manner

• performing maintenance on a regular, predictable schedule so workers know when to expect it

• Determining the best time to perform preventive maintenance on equipment

• Determining when to outsource maintenance

Predictive maintenance is used to determine:

• when preventive maintenance was last performed

• when preventive maintenance should be performed

• which resources should be maintained via preventive maintenance

• which resources should be maintained via breakdown maintenance

1. Which of the following would not typically be a consideration in the equipment replacement question?

• Forecasts of future demand

• Expertise workers had on the old equipment

• The cost of the replacement equipment

• The cost to maintain the old equipment

Some of the major benefits of a smart factory?

• Boosting Overall Equipment Effectiveness (OEE) only

• Boosting Overall Equipment Effectiveness (OEE) and Flexible, adaptive, and proactive production

• Boosting Overall Equipment Effectiveness (OEE), Flexible, adaptive, and proactive production and Real-time, on-demand visibility into performance across the production chain

• Reduce manpower