In almost 600 BC, the Green Walls concept was presented. Most of the previous research has proved that the use of green walls in different types of interior spaces has its considerable merits in improving users' overall health and productivity. Interior Designers seem reluctant to integrate green walls in underground spaces due to its drawbacks as one of the most expensive man-made walls, needing scheduled water requirements, and its susceptibility to adversities such as fungi growth.

The current study offers an intelligent solution predicting the performance of self-sustainable green wall systems improving underground air quality. The system is mitigating the issue of the green wall's short life span. The sustainability of green walls and air quality in underground interior spaces is investigated by applying IoT and AI technologies.

Results of the present work show that different Random forests which an example of an ensemble learner built on decision trees. The Decision trees are extremely intuitive ways to classify or label objects models were generated compared and evaluated for accuracy and sensitivity. These Models were built simulating IOT-based-Air quality monitoring systems integrated with self-sustainable green walls. The Datasets selected included features like temperature close to that of the targeted public interior spaces like underground stations in Egypt. Predicting underground air quality has been conducted by indicating pre-defined parameters such as PM and CO.
These results can evolve in the near future, enabling decision-making systems to predict the performance of similar self-sustainable multi-purpose green walls in maintaining underground space air quality.