Customer Service Automation
The group needed to support a very large water-service operation, but staff were limited, service coverage was inconsistent, knowledge quality varied across workers and customers could not get help around the clock.
Reported impact
The utility created a 24x7 digital service layer that intercepts routine inquiries, improves staff consistency and gives managers a closed-loop view of service risk and resource allocation.
Shenzhen Environment & Water Group and Baidu AI Cloud built the 'Shenxiaoshui' intelligent agent on top of Baidu's Qianfan platforms and the group's existing Shenshui Cloud Brain. The deployment is positioned as the country's first water-service agent and creates a three-sided service model for citizens, field water stewards and managers. On the citizen side, the agent simulates a human service representative and handles supply and drainage consultations 24x7. For field staff, it breaks through data silos across business systems so they can retrieve user consumption, work-order status and standardized response knowledge quickly. For managers, the agent aggregates service data, monitors performance indicators and supports a warning-dispatch-analysis-iteration loop that recommends service strategies and generates multidimensional analysis reports. All of this runs across a large operational footprint that includes hundreds of service workers, thousands of communities and millions of users, making the agent both a front-line service interface and an internal decision-support layer.
Customer questions, user water-consumption records, work-order states and service metrics flow into Shenshui Cloud Brain and Qianfan-backed agent workflows; Shenxiaoshui then returns answers to citizens, knowledge and lookup results to stewards, and risk warnings and analysis reports to managers.
Baidu AI Cloud and Shenzhen Environment & Water Group built the country's first water-service intelligent agent, 'Shenxiaoshui,' on Qianfan AppBuilder and Shenshui Cloud Brain (original: 依托百度智能云千帆AppBuilder平台以及“深水云脑”，构建了国内首个水务服务智能体“深小水”).
It can quickly retrieve information such as user water consumption and work-order status, with 90% of query needs answered in time (original: 快速调取用户用水量、工单状态等信息，90%查询需求及时响应).
It provides 7×24 uninterrupted service and can intercept 50% of routine consultations on the intelligent side (original: 为用户提供7×24小时不间断服务……将50%常规咨询拦截在智能端).
The agent covers 459 water stewards, 7,160 communities and 3.22 million users (original: 覆盖全市459名水管家、7160个小区、322万用户).
The management workflow builds a 'warning-dispatch-analysis-iteration' closed loop (original: 构建“预警-调度-分析-迭代”闭环).
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 Sep 24, 2025 All-China Federation of Trade Unions The All-China Federation of Trade Unions needed a unified AI foundation that could support legal aid, rights protection, and other worker services at national scale instead of leaving union service digitization… Delivery cycle fell from 3-6 months to under 1 month. Qianfan AppBuilder · Qianfan ModelBuilder · Union Industry Model Baidu AI Cloud Sample Jun 19, 2025 Xi'an Public Security Bureau Traffic Management Department (西安市公安局交通管理局) Xi'an traffic police needed a scalable way to handle rising traffic-service inquiries, relieve hotline pressure, remove information silos between frontline officers and functional departments, and standardize both… Answer accuracy exceeded 96%. Qianfan AppBuilder · Qianfan ModelBuilder · Yong'anjun Baidu AI Cloud Sample May 15, 2025 Zhongshan 12345 Government Service Hotline The Zhongshan 12345 hotline had to serve thousands of daily calls across dozens of agencies, but knowledge was fragmented, address entry and work-order creation were manual, and city-event analysis was too slow to move… 90 operators support the city's requests. All-Elements Governance Engine · Smart Analysis and Early Warning · Smart Work-Order System 