Zong says the congestion-fixing work that used to take its engineers days is now done by software on its live network. The operator’s Self-Intelligent RAN Optimization Platform spots a crowded cell, applies a correction and then checks whether the fix worked, with no person in the loop.
The Fast Mode Awards 2026 named it winner of its Operational Excellence Leader prize in the innovation category. Zong says it is the first time anyone in Pakistan has run Level 4 autonomous operations on a network carrying real customers, and that its own engineers built it. It is the company’s second international award this year for AI work.
For subscribers, the practical promise is steadier service in busy areas, because slow cells get retuned faster than a manual cycle allows. Zong gives no figures, though: no cell counts, no cities, no before-and-after speeds, and no date by which it expects the whole network to run this way.
Chief technology officer Mao Weiliang said the company is “taking AI beyond monitoring and recommendations to autonomous decision-making and execution”. He framed it as groundwork for networks that tune themselves across the whole country.
Radio access network optimization is the routine tuning of antennas, power and capacity settings so that crowded cells do not drag down everyone connected to them. Zong says the old process was manual and could take several days from spotting a problem to confirming it was gone, which is the gap the platform is meant to close. The operator also says the system is built to scale, so the same loop can be applied across more of the network over time.
Zong is a subsidiary of China Mobile and sells its 5G service under the “Aisa Hai Future” tagline.
This story is based on a press release from Zong.
