The monitoring of the Internet of things networks depends to a great extent on the
availability and correct functioning of all the network nodes that collect data. This network
nodes all of which must correctly satisfy their purpose to ensure the efficiency and high quality
of monitoring and control of the internet of things networks. This paper focuses on the problem
of fault-tolerant maintenance of a networked environment in the domain of the internet of things.
Based on continuous-time Markov chains, together with a cooperative control algorithm, a novel
feedback model-based predictive hybrid control algorithm is proposed to improve the maintenance
and reliability of the internet of things network. Virtual sensors are substituted for the sensors that
the algorithm predicts will not function properly in future time intervals; this allows for maintaining
reliable monitoring and control of the internet of things network. In this way, the internet of things
network improves its robustness since our fault tolerant control algorithm finds the malfunction nodes
that are collecting incorrect data and self-correct this issue replacing malfunctioning sensors with
new ones. In addition, the proposed model is capable of optimising sensor positioning. As a result,
data collection from the environment can be kept stable. The developed continuous-time control
model is applied to guarantee reliable monitoring and control of temperature in a smart supermarket.
Finally, the efficiency of the presented approach is verified with the results obtained in the conducted
case study.
