Abstract: In this paper, we present several considerations centeredaround the data-driven system approaches. We briefly explore threemain issues: the evolving relationship between off-line and on-linedata processing methods, the complementary relationship between thedata-driven and model-based methods, and the perspectives ofdata-driven system approaches. Instead of offering solutions todata-driven system problems, which is impossible at the presentlevel of knowledge and research, in this article we aim atcategorizing and classifying open problems, exploring possibledirections that may offer alternatives or potentials for the four key fields of interests: control, decision making, scheduling, and fault diagnosis.