Smart Factory Fundamentals and Industrial IoT
This domain tests whether you can separate a smart-factory architecture from generic IoT, place functions at the correct ISA-95 level, and reason quantitatively about OEE and IIoT data flow.
Core architecture
- ISA-95/Purdue levels: 0 sensors and actuators, 1 control, 2 supervisory, 3 operations/MES, 4 business/ERP.
- IT/OT convergence is a trade-off: OT values determinism, availability, and safety; IT values confidentiality, elasticity, and scale.
- A typical IIoT path is sensor → edge gateway → edge compute → historian or cloud. Store-and-forward is mandatory for intermittent links; timestamps should be created as close to the source as possible.
- OEE = Availability × Performance × Quality. Availability uses planned production time and excludes planned stops. Performance uses ideal cycle time; quality uses good count over total count.
Worked example
A line is scheduled for 480 minutes, with a planned 30-minute break, 45 minutes of unplanned downtime, 700 units produced, 14 defects, and an ideal cycle time of 0.5 minutes per unit. Planned production time = 450 min; run time = 405 min; availability = 405/450 = 0.90. Ideal run time = 700 × 0.5 = 350 min; performance = 350/405 = 0.864. Quality = 686/700 = 0.98. OEE = 0.90 × 0.864 × 0.98 = 0.762, or 76.2%. Notice that averaging the three ratios would give a different, incorrect result.
Traps that cost marks
Averaging OEE components instead of multiplying them; including planned breaks in availability; using actual cycle time in performance; treating MQTT alone as a complete IIoT architecture; and ignoring edge buffering, time synchronization, or data ownership.
Key points
- Know the ISA-95 levels and what belongs at each.
- OEE is multiplicative and uses planned production time.
- OT constraints differ from IT constraints.
- Edge buffering and source timestamps protect data integrity.