Knowledge Networks For Decision Making in Infrastructure Asset Management

管理我们的城市基础设施系统决策变得越来越复杂。一项研究发现，在一个城市，水管理系统跨越200个业务流程。在安大略省，2010年至2014年间，环境评估过程的成本上升了340%。平均持续时间从19个月跃升至27个月。根据一项研究，如果业主使用先进的工具并采用最佳实践，可以节省大约40%的基础设施建设支出。******本提案的目标是研究，建模和开发一个环境，用于使用项目网络的社会语义网络分析，以在决策系统中注入最佳实践。项目网络（PN）包括决策者和公民的社会网络、项目文件网络、BIM（建筑信息模型）所代表的物理项目元素网络、它们的成本网络（预算）和活动网络（进度）。***通过参与实际项目，学生将建立至少10个案例研究。首先，他们将通过调查获取每个项目的关键属性，例如每个项目的实际社会网络（谁与谁沟通）、决策路径、参与者影响力水平。他们还将（通过Twitter界面）收集公民在社交媒体上的辩论。学生将收集专题文件，形成概念的语义网络。他们将调查项目团队，以确定最重要的主题、绩效水平，以及对两者的看法。***其次，学生将使用相关的社会和语义网络分析算法，为每个项目提取相同关键属性的指标。***将比较两组关键属性（来自调查和算法），这可以帮助定义社会和语义网络分析算法对基础设施领域的适用性。******通过案例研究和调查的比较分析和算法分析，学生可以开发模型：***形成子群体或已知的兴趣社区：目的是对比正式和非正式的决策路径。***意见动态：衡量利益相关者（包括当地社区）对主要项目主题的立场和情绪的变化。***信任分析：监测主要项目参与者之间的信任形成情况。PN本体：列举适当的算法及其在基础设施系统中的潜在用途。***BIM仪表板：开发一个门户，将PN分析结果嵌入到项目管理实践中。******这项研究工作的贡献在于充分理解了PN，以及如何将从分析中获得的知识融入决策周期。这有助于1)提高社区参与度，2)加强决策过程，3)降低成本。所有项目参与者都能从成果中受益，尤其是政府官员和项目决策者以及社区成员。

Managing our urban infrastructure systems decision making is increasingly becoming complex. A study found that in one municipality, the water management system spans 200 business processes. In Ontario, the cost of environmental assessment process rose 340% between 2010 and 2014. Average duration jumped from 19 to 27 months. According to one study, about 40% of expenditure in infrastructure construction can be saved if owners utilize advanced tools and deploy best practices. ******The goal of this proposal is to study, model and develop an environment for using social-semantic network analysis of project networks to infuse best practices in the decision making system. Project networks (PN) encompass the social networks of decision makers and citizens, networks of project documents, networks of physical project elements as represented in BIM (Building Information Modeling), their cost network (budget) and activity network (schedule). ***Students, through engagement in actual projects, will build at least 10 case studies. First, they will capture, through surveys, key attributes of each project—such as its actual social network of each (who communicates with who), decision paths, levels of actor influence. They will also collect (through Twitter interface) social media debates by citizens. Students will collect project documents, which form a semantic network of concepts. They will survey project teams to define top topics, performance levels, and the sentiment regarding both. ***Second, students will use relevant social and semantic network analysis algorithms to extract indicators for the same key attributes for each project. ***The two sets of key attributes (driven from surveys and from algorithms) will be compared, which can help define the suitability of social and semantic network analysis algorithms to the infrastructure domain. ******Through case studies and comparative analysis of surveys and algorithmic analysis students can develop models for:***Formation of sub groups or what is known communities of interest: the aim is to contrast formal and informal decision paths. ***Opinion dynamics: to measure changes of stakeholders' (including local communities) position and sentiment regarding main project topics. ***Analysis of trust: monitor trust formation between major project players. ***Ontology for PN: enumerating appropriate algorithms and their potential usage in infrastructure systems.***BIM dashboard: develop a portal to embed the results of PN analytics into the practice of project management. ******The contribution of this research work is full understating of PN and how to ingrain knowledge gained from their analysis into decision making cycles. This can help in 1) improve community engagement, 2) enhance decision making process, 3) reduce costs. All project participants can benefit from the results—particularly public officials and project decision makers as well as members of communities.