AI APPLICATIONS IN SIX SIGMA: INTEGRATING MACHINE LEARNING INTO DMAIC FOR TELECOMMUNICATIONS PROCESS IMPROVEMENT

Six Sigma has become one of the most powerful data-driven process improvement methodologies and has
provided organizations with a structured method on how to reduce variation and enhance quality. The DMAIC
cycle of Define, Measure, Analyze, Improve and Control has long been a potent source of systematic problem
solving and operational excellence as Bukhari and Akhtar (2024) explain. However, the contemporary industrial
and service environment has greatly transformed, and the traditional Six Sigma tools are growing to be questioned
by the magnitude, velocity, and intricacy of the current data environments.
The transition is particularly evident in the industries that have high-volume and high-velocity data in their
operations, like the telecommunications. Channappagoudar, Bhat and Gijo (2025) claim that telecom systems are
producing continuous data streams of network gear, customer interactions and digital interfaces, which puts an
analytical load much larger than that of the original design of the Six Sigma techniques. Industry 4.0 technologies,
as these authors observe, are changing the quality improvement expectations, and organizations have to
incorporate intelligent, automated, and predictive into their DMAIC processes.
Although the actual change is still in the realm of Artificial Intelligence (AI) and Machine Learning (ML), these
two elements are gaining popularity as the facilitators of this process. According to Kabeer (2025), AI, specifically
machine learning, can be used to assist in continuous process improvement as it can help find patterns and process
anomalies that could otherwise be challenging to identify. On the same note, Sood and Dhull (2024) affirm that
AI compliments the Six Sigma in bolstering predictive analytics, therefore, improving the ability of DMAIC to
foresee defects, variability and performance concerns before they become aggravated.
The concept of AI inclusion in Six Sigma belongs to an active trend of the predictive, digital quality management.
Santacruz (2023) emphasizes the fact that integration of ML with the Six Sigma helps organizations to transition
to zero-defect performance by changing problem solving to simulation and proactive forecasting. Similarly,
Maged, Haridy, Awad and Shamsuzzaman (2023) present the actual application of machine-learning-supported
Six Sigma tools, indicating that AI improves the flexibility and analysis depth of the DMAIC cycle in various
working scenarios.