Smart Factory Data Analytics Flashcards
7 cards from real SACA practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Smart Factory Data Analytics flashcards as text
Which protocol is most commonly used to transport IoT sensor data from factory floor devices to a broker in a smart factory?
Answer: MQTT
MQTT is a lightweight publish-subscribe protocol designed for constrained IoT devices and low-bandwidth, high-latency networks typical in factory environments.
A factory analyst wants to identify which combination of machine parameters most frequently precedes a product defect. Which analytical approach is most appropriate?
Answer: Association rule mining on process parameter logs
Association rule mining discovers frequent co-occurring parameter combinations in event logs, revealing multi-factor patterns that precede defects.
What is 'data drift' in the context of a smart factory ML model deployed in production?
Answer: A gradual change in the statistical distribution of incoming data compared to training data
Data drift occurs when the real-world input distribution shifts over time, causing a model trained on older data to lose prediction accuracy.
In a smart factory dashboard, a line chart shows sudden spikes in cycle time every Monday morning. What is the most likely root cause to investigate first?
Answer: A weekly scheduled maintenance or shift changeover event
Periodic spikes aligned with a calendar pattern (Monday mornings) strongly suggest a recurring scheduled event such as maintenance, startup, or shift handover.
Which technique reduces the dimensionality of high-dimensional sensor data while preserving the most variance, aiding visualization and model training?
Answer: Principal Component Analysis (PCA)
PCA projects high-dimensional data onto a lower-dimensional space that captures the maximum variance, making it easier to visualize patterns and train models.
A smart factory collects temperature readings from 500 machines every second. After 30 days, the raw dataset is approximately how large if each reading is a 4-byte float?
Answer: ~5.2 GB
500 machines × 1 reading/sec × 86,400 sec/day × 30 days × 4 bytes ≈ 5.18 GB, illustrating the data volume challenges in industrial IoT.
What is the role of a message broker like Apache Kafka in a smart factory data pipeline?
Answer: It acts as a durable, high-throughput buffer between data producers and consumers
Kafka decouples producers (sensors, PLCs) from consumers (analytics engines, databases) by buffering messages durably at high throughput.