The evolution of Smart Cities conveys continuous changes involving a great number of variables, which might hamper the development of evaluation tools and methodologies. Most of the metric models for Smart City are based on the selection of key performance indicators (KPI) according to the specific model objectives. As different organisations propose their own indicators generating different models, it is difficult to get a straightforward comparison among models. With the aim of dealing with this and other disadvantages, in this study, a framework based on the application of Data Science to the KPIs is proposed. This framework represents an infrastructure that goes through the treatment of Open Data, facilitating the evaluation of different models comparison intended for decision-making, and to the final stage of dispatching service reports. There are four components that integrate this framework (i) a tree structure to manage the KPIs; (ii) a designed JavaScript Object Notation document for service dispatch; (iii) Web applications for evaluations based on Smart People with four scenarios and; (iv) the infrastructure for reception and attention of reports.

中文翻译：

---

支持面向有效上访服务请求的决策的智能城市模型的数据科学框架

智慧城市的演进传达出不断变化的变化，涉及许多变量，这可能会阻碍评估工具和方法论的发展。智慧城市的大多数度量模型均基于根据特定模型目标选择的关键绩效指标（KPI）。由于不同的组织提出了自己的指标来生成不同的模型，因此很难在模型之间进行直接比较。为了解决这个和其他缺点，本研究提出了一个基于将数据科学应用于KPI的框架。该框架表示一个基础结构，该基础结构经历了开放数据的处理，有助于评估用于决策的不同模型的比较，并达到分发服务报告的最后阶段。有四个组件集成了此框架（i）树形结构来管理KPI；（ii）设计用于服务分配的JavaScript对象符号文档；（iii）基于智能人的评估用Web应用程序，具有四种方案；以及 （iv）接收和关注报告的基础设施。