management, errors are inevitable with COBie drops. Meanwhile, a portion of COBie data is required to be populated and updated manu - ally. This creates more trouble over the long run because automated population must be maintained along with manual inputs. Most firms do not prepare and identify this as a risk in the project data manage- ment phase, which happens to be the most critical part of BIM to FM lifecycle implementation. Infrastructural Unfriendly COBie cannot handle infrastructure projects like other coding standards (i.e., UniFormat™). This is a major drawback, and even an immedi - ate turn off, for Transit Agencies, DOTs, Port Authorities, which own most of the infrastructural capital projects like bridges, tunnels, piers, highways, as shown in Figure 2. For most projects, these agencies will start to develop their own asset structure system rather than use COBie, UniFormat™, or OmniClass®. For example a transit agency owns various types of capital projects in all the above mentioned infrastructural categories. UniFormat 2010 has been adopted in their project delivery phase and is regulated by all con- tractual languages, BIM standards, and BIM execution plans. However, the facilities have their own coding system due to the fact UniFormat™ is a very building-centric system and pushes them to develop their own AM/FM centric data structure. The data structure asymmetry creates a gap from PIM to AIM. An extra mapping layer is required to bridge the gap in terms of data validation, migration, and transition. Make it More Complicated COBie can be useful, rational, and practical if only every stakeholder has agreed, is committed and they all have a good understanding of data exchanges among different stages and drops. Some may already find this is almost an impossible task from a strategic perspective and the need to prepare everything at the beginning of the project, with every stake- holder sitting at the same table, defining data requirements, and COBie schema. That does not include the BIM Standard, BIM Execution Plan (BxP), and QA/QC Compliance Checking process which all need to be correspondingly updated to function with the COBie implementation. The complexity in collaboration, implementation, and validation further demotivates most from choosing COBie in the first place.
common “database” between all stakeholders. However, COBie is still confusing most of the time even though it aims to be the “chosen one” to solve all the data problems during the BIM lifecycle implementa- tion. And we all know even with the same last name, chosen ones can have dramatically different stories. We will discuss COBie’s story of BIM2FM transition in the next section. Real World Complications The AECO industry has high expectations for COBie to be the data savior. Anyone ever involved in COBie-enabled projects remembers the size of the COBie spreadsheet. COBie can easily generate inten - sively long spreadsheets due to the number of attributes of every object exported. In extreme scenarios such as an airport terminal, large-scale hospital, or higher-ed facility, it will exceed Excel’s workspace limit of around one million rows. With an acceptable error rate of 1 percent to 1.3 percent published by buildingSMART industry tests, this will lead to 10,000 rows of problematic data and a painful process consider - ing the difficulty of eyeballing and manually correcting an oversized, color-coded spreadsheet, even on a powerful workstation. The Struggle of Data Errors COBie is prone to data errors and that is another barrier stopping companies from adopting it on a lifecycle level. BIM embraces an extensive portfolio of applications, processes, systems, involving various stakeholders, vendors, trades, etc. Data from all channels need to be mapped to populate the COBie sheets – which is more like a highly-maintained business process, rather than a pure technical data exchange. Based on experience with COBie-streamlined lifecycle data Figure 2: Illustration of Weak Capability of Handling Non-building Infrastructural Projects.
Figure 3: Similarity of UniFormat 2010 and Animal Taxonomy
42
csengineermag.com
september 2020
Made with FlippingBook Annual report