Beyond the Building: Urban Information Modeling (UIM)

Beyond the Building: Urban Information Modeling (UIM)

Project Team

Rishee Jain, Zheng Yang, Andrew Sonta

Overview

Analysis and design on the building scale takes place in parallel with that on the urban scale, creating difficulty in understanding how the building fits in with the larger city context early in the design process. Such misalignment can result in delays and cost overruns (~5.7%) during construction of a building.

An Urban Information Model (UIM) that is able to capture the various urban data streams, aspects of a regional plan set by agencies, and information from the virtual design of a building would enable better and earlier decision-making with regard to aspects of the project that lead to delays later in the project life cycle. Structured analysis will be conducted to classify various urban data streams. A method to combine such data in an ontological framework (UIM) will be developed and validated on a case study with a CIFE industry collaborator.

Project Background

Research Motivation

More people than ever are living and working in cities. As the share of the global population living in urban areas grows to 66% by 2050, planners, architects, and builders must strengthen their understanding of the complex dynamics of city life. The many design and construction decisions that go into interventions in the built environment — e.g. buildings, infrastructure projects, or policy decisions — would be better made if they are informed not only by the goals of the owner, conditions of the site, and coordination among stakeholders, but also if they are informed by their potential impact on the surrounding urban environment. To work toward this goal of a robust understanding of the city’s form and its occupants, we must make use of the myriad urban data streams that are continuously being produced in the city.

The urbanist Jane Jacobs’s description of the kind of problem a city is — her “organized complexity” — is beginning to reveal itself through technologic advancements that have increased the availability of the unstructured and structured data describing the urban world. From three dimensional spatial models of the built environment, to traffic patterns, to socioeconomic mappings, urban data is being collected at different scales, by different people, in different time intervals, and through many different systems. For the most part, the current way we think about city-scale data is specific rather than comprehensive: traffic data is used to solve traffic problems; economic data is used to solve economic problems. With all of this fragmented information that describes city life, it is difficult to identify and understand the relevant multi-dimensional effects on our urban environment that come from city-scale interventions.

Industry Example

In modern-day construction, delays caused by unknowns can contribute heavily to increases in costs. One common reason for a project delay is an environmental assessment followed by an environmental impact statement. Chester and Hendrickson found that such critical delays can lead to project cost increases of up to 5.7%. For a project as large as the Salesforce Tower in San Francisco, with an estimated budget of $1.1 billion, a delay for re-assessing environmental air quality or pedestrian flow could have huge cost implications for the project (e.g., $62.7 million).

Research Objectives

The proposed research will consist of the following three steps:

  1. Analysis and classification of the various data and information streams that are already available within urban areas, as well as those becoming more readily available.
  2. Development of a method for combining these data streams in an elegant ontological framework (i.e., core of the Urban Information Model) that allows for straightforward representation of the many dimensions of the urban environment.
  3. Validation of the ontological framework on a case study in collaboration with a CIFE industry collaborator.

Expected Results

The biggest anticipated outcome of this research is a novel and elegant multi-dimensional Urban Information Modeling framework that will allow for the synthesis of the many data streams available in cities. This framework will be an ideological extension of the BIM ontology at a larger urban scale. The research team is not aware of an existing framework that works to synthesize the data describing the spatial, temporal, and human dimensions of a city.

To be able to simultaneously analyze various data streams will enable decision-makers to understand impacts that result from policy recommendations or new proposals for urban systems. With this proposed work, the ontological framework of BIM can move beyond design and construction of buildings, and create a new framework that will enable contractors, urbanists, sociologists, policy makers, and architects to leverage integrated engineering tools for better design and analysis of city-scale interventions.

Project Update

Over the last several months, we have accomplished the following tasks related to our project:

  • Civic-Academic Collaboration: Established a collaboration with City of Palo Alto’s Chief Information Officer (CIO) – Dr. Jonathan Reichental.
  • Case Study Site: Identified the case study site in Palo Alto for development of the UIM framework (University Avenue corridor).  Began administrative and legal process for acquiring existing data and deploying new sensors at the study site.
  • Problem Identification:  We further identified and clarified the problem setting by interviewing several data analysts and other officials at the City of Palo Alto.
  • Data Classification:  In conjunction with the City of Palo Alto, we explored the various data and information streams currently being collected by the city and classified the streams by spatial and temporal granularities.
  • New Ontology/Framework: Developed a data integration and ontological framework (Figure 1) based on geo-relationship to manage urban data (e.g., buildings, roads, trees, and sensing systems) with different time scales and variant availabilities from heterogeneous sources. It specifically derives the geo-relationships of urban data by utilizing available geospatial information and automatically learning the ontologies (e.g., classification of relationships). This framework could effectively integrate urban data without introducing any application specific bias or assumption (e.g., energy network modularity), which provides a fair and consistent basis for the objective of the UIM to analyze and quantify the interactions and interdependencies of various urban systems (e.g., building performances, traffic conditions, and micro climate). Based on the developed data integration framework, the urban data are reconstructed and structurally stored in a graph database instead of a traditional relational database in order to enable the framework to handle spatially and temporally heterogeneous data, improve computational run time and maximize usability by city officials.

  • Conference paper: Disseminated our initial results (all previous tasks) through a paper and oral presentation at the 2017 International Workshop on Computing in Civil Engineering (IWCCE 2017) in Seattle, Washington.  The full text of this paper is provided below.

Original Research Proposal

AttachmentSize
UIM_CIFE_Jain_Lepech_Sonta_vF.pdf648.94 KB
IWCCE - A data integration framework for urban systems analysis based on geo-relationship learning.pdf991.35 KB

Last modified Tue, 27 Jun, 2017 at 13:59