Table 22 Real-World testbed results over a 48-hour smart-factory workload.

From: NF-MORL: a neuro-fuzzy multi-objective reinforcement learning framework for task scheduling in fog computing environments

Metric

NF-MORL

HTSFFDRL

DDPG-TS

GA-TS

Average makespan (s)

142.3

208.7 (+ 46%)

231.4 (+ 62%)

265.1 (+ 86%)

Energy consumption (kWh)

9.81

13.92 (+ 42%)

14.67 (+ 50%)

16.34 (+ 67%)

Monetary cost ($)

2.41

3.58 (+ 48%)

3.91 (+ 62%)

4.27 (+ 77%)

Task success rate (%)

95.8

81.2

77.6

69.3

Back to article page