IT Operations Automation · Real-time Analytics
After Typhoon Yagi damaged most field sensing equipment, Haikou traffic police urgently needed to restore signal-control effectiveness without waiting for every radar, camera, and detector to be rebuilt, or the city would remain stuck with outdated fixed-time traffic signaling.
Reported impact
Haikou restored practical traffic-signal optimization quickly after the disaster, reduced pilot-intersection delay, and created a lower-cost operating model that could support day-to-day signal management even before full hardware recovery.
Haikou Traffic Police and Baidu AI Cloud rapidly deployed a signal-control SaaS platform to recover urban traffic optimization after severe typhoon damage. Before the disaster, the city had built a video-plus-radar monitoring system with edge-computing units, video detectors, and radar detectors. When the storm damaged 80% of field sensing equipment across hundreds of intersections, Baidu introduced an upgraded SaaS control platform that packages signal-plan identification, diagnosis, optimization, and reporting into a lightweight service. The deployment used 60 downtown pilot intersections and combined internet floating-car data with traffic-flow feature data, plus algorithmic reconstruction and calibration, to rebuild traffic volumes even when normal roadside collection devices were missing.
Roadside sensing data, floating-car internet data, and traffic-flow feature data are fed into algorithmic reconstruction and calibration workflows. The SaaS platform then generates signal timing analyses, optimization plans, and reports that traffic police validate and use for intersection operations.
Haikou traffic police urgently launched a signal-control SaaS platform with Baidu to recover smart-traffic operations after the typhoon (original: 海口交警紧急联合百度智能云，推出了SaaS信控平台).
The city had 567 affected intersections and 80% of field sensing equipment was damaged (original: 当晚11点全市共有567个交通路口受到影响，80%外场感知设备受损).
The pilot used 60 complex downtown intersections and reconstructed flow with 'internet data + algorithm models' plus floating-car and traffic-feature data (original: 项目选择了中心老旧城区的60个路口作为试点…基于“互联网数据+算法模型”的方法，结合互联网浮动车数据、交通流特征数据，进行了交通流量的数据还原与校准).
The platform kept flow and direction error within 20% and reduced average per-vehicle delay by 14.37% at pilot intersections (original: 确保流量和流向数据误差控制在20%以内…试点路口，车均延误时间更是下降了14.37%).
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
Baidu AI Cloud Sample May 14, 2025 Jingxiong Expressway Hebei Section Jingxiong Expressway needed a more precise and collaborative operating model because fragmented systems, weak video perception, rigid workflows, and slow emergency coordination limited how effectively highway managers… Business-system efficiency is expected to improve 80%+. BaiGe Heterogeneous Compute Platform · Highway Industry Large Model · Qianfan Large Model Platform Baidu AI Cloud Sample May 13, 2025 Hebei Expressway Group Jingxiong Expressway Branch (河北高速公路集团有限公司京雄分公司) Jingxiong Expressway operators struggled with rigid workflows, fragmented data and systems, weak video perception, and slow emergency handling, which limited the precision and responsiveness of highway operations… Business-system efficiency is expected to improve 80%+. Highway Industry Large Model · Qianfan AppBuilder · Qianfan ModelBuilder Baidu AI Cloud Sample May 8, 2025 Haidian District (北京市海淀区) Haidian's rapid urban growth made manual, department-siloed governance too slow and expensive, especially for problems like muck-truck regulation where data fragmentation and cross-department coordination prevented… 160,000 data items processed daily Data Service Center · Haidian City Brain 