In this study, we explore the benefits of employing an AI-based control system to enhance the optimization response and establish an AI bot capable of conducting pre-design traffic analysis through data acquired over its operational lifespan.

Since the 1960s, lift control systems have transitioned from relay-based to utilizing neuro-fuzzy algorithms, driven by advancements in computational power and electronic technology, focusing on optimizing car/hall allocation, waiting times, and loading capacity.

The primary challenge with current systems is their inability to adapt to unforeseen changes in traffic patterns, building usage alterations, and modifications made to the building's setup after the initial design phase or during its lifetime.

Additionally, changes in user behaviour and unforeseen events in the building can lead to unexpected lift system saturation.

To enhance service quality, it is clear that implementing an AI-driven decision-making process is essential.

The key difference between intelligent systems and static ones lies in the capability of intelligent systems to self-improve based on future predictions made using AI and machine learning techniques.

The approach involves training a standalone AI engine with the broadest set of real data available for training to create an optimization system for elevator management.

Incorporating a Natural Language Layered Model (NLLM) into the AI engine allows system conditions to be negotiated through linguistic means. For example, a building manager could set parameters for the AI, such as instructing the system to request passengers on the first two floors to use stairs instead of elevators on busy Saturday evenings, thereby modifying the algorithm controlling the elevator system accordingly.