With the recent development and applications of the Internet of Things (IoT) paradigm, smart home concepts are prominent in providing reliable monitoring of people through remote access. The information and communication technologies are harmonized for providing comfortable assistance for elderly and home-cared peoples. The reliability of such smart home-based IoT networks is defaced due to multiple concurrencies in automation control systems. This concurrency issue is resolved by designing a novel pivotal complementary automation control model (PCACM) using a machine learning paradigm. This model first observes the changes and requirements in the smart home environment. The decision tree-based learning paradigm constructs non-congruent solutions for providing different levels of concurrency. By identifying the fixed and varying control routines, the tree is constructed with the possible decisions. If the control requirement exceeds the available solutions, then complementary actions are assimilated, and the tree is modified. Such a decision-making process can confront older adults’ ad hoc requirements, regardless of time and benchmarked assistance. The proposed model’s performance is verified using the metrics assistance attainment, delay, control failures and overhead.

随着物联网 (IoT) 范式的最新发展和应用，智能家居概念在通过远程访问提供可靠的人员监控方面显得尤为突出。协调信息和通信技术，为老年人和家庭护理人员提供舒适的帮助。由于自动化控制系统中的多重并发，这种基于智能家居的物联网网络的可靠性受到损害。这个并发问题是通过使用机器学习范式设计一种新颖的关键互补自动化控制模型 (PCACM) 来解决的。该模型首先观察智能家居环境的变化和需求。基于决策树的学习范式构建了非全等解决方案以提供不同级别的并发性。通过识别固定和变化的控制程序，树是用可能的决定构建的。如果控制要求超过了可用的解决方案，则补充行动被同化，并且树被修改。这样的决策过程可以面对老年人的临时要求，而不管时间和基准援助。使用度量辅助实现、延迟、控制失败和开销来验证所提出模型的性能。