随着配水基础设施的老化，公用事业管理人员 必须权衡管道维修的效益和成本 替代品。尽管已经提出了一些模型 为了帮助管理者做出这些决定，他们无法 结合环境和地理因素 影响基础设施状况。本文介绍一种诊断方法 使用系统数据和管道破裂记录提供帮助的工具 管理者识别网络中包含攻击性攻击的区域 最容易发生故障的环境。 攻击性指数（AI）是基于历史数据的指标 网络性能。空间分解用于评估 网络内的可变性；观察到的断裂数据用于推断 跨域的可变攻击性。人工智能可以实现电流和电流 应考虑攻击性环境的潜在影响 无需明确识别和测量物理特性 这可能会影响铸铁管网的退化。 需要制定有效的管道更换计划 如果不深入了解当前和未来的经济状况，这是不可能的 恶化对于管道故障率较低的系统，网络 对影响因素的监测或量化几乎是不可能的 和管道更换一样昂贵。人工智能提供了另一种选择 采用成本高昂的基于监控的方法更换主管道，并 使公用事业管理器能够编译高效的替换 以特定区域的管道为目标的明细表。 包括26个参考文献、表格和图表。

As water distribution infrastructure ages, utility managers must weigh the benefits and costs of pipe repairs against those of replacement. Although models have been proposed to help managers make these decisions, they have not been able to incorporate the environmental and geographic factors that influence infrastructure condition. This article introduces a diagnostic tool that uses system data and pipe break records to help managers identify regions across the network that contain aggressive environments and that may be most prone to failure. The aggressivity index (AI) is an indicator based on historical network performance. Spatial disaggregation is used to assess variability within the network; observed break data are used to infer variable aggressivity across the domain. The AI enables current and potential influences of aggressive environments to be factored in without the need to explicitly identify and measure physical properties that may influence degradation of cast-iron pipe networks. Development of an effective pipe replacement schedule is impossible without insight into the current and future state of deterioration. For systems with a low rate of pipe failure, network monitoring or quantification of influential factors may be almost as expensive as pipe replacement. The AI provides an alternative to costly monitoring-based approaches to main replacement and enables utility managers to compile an efficient replacement schedule that targets pipes in specific areas. Includes 26 references, table, figures.