I am sure that we have all had experience with collaborative work on a project, and the complications which arise when cross-discipline work is required. BIM programs such as Revit have allowed companies to better coordinate their projects, but these programs still have quite a few limits. For example, often BIM models are used only for the coordination of geometry. Calculations are instead done in separate programs, with each division forming separate models in addition to the central model. Updates then often need to be done manually, which may lead to errors. An increases in beam size which is not notices in the structural department may not be immediately updated in the Revit model, for example, and this may lead to coordination problems when a mechanical engineer places a duct or piece of equipment where he or she believes there is space. While the capabilities of BIM to help coordinate projects are incredible, such problems do exist.
Chapter Two in Charles M Eastman's BIM Handbook covers interoperability, the ability of programs to work together. In the above example, the structural software was not able to update the BIM model itself, and users must check for changes and update models. This leads to the possibility of human error in translation. Eastman discusses the various methods for communicating between programs, including file types such as .dxf which can be read and produced from multiple programs, as I am sure at least a few of us have experienced. I personally have used this capability to form AutoCAD .dwg files from structural models in the past, but this is only a line model and does not directly transition into a BIM model.
The section which most held my interest in the chapter was the discussion beginning on page 114 about Industry Foundation Class (IFC). This representation of various parts in a BIM model is able to coordinate various properties and definitions of different parts of a model. For example, a wall is associated with various spaces it is adjacent to, as well as its materials, fire rating, purpose, etc. While each wall may have many properties, an engineer may be interested in only a few. The structural engineer may not care about the thermal transmittance, for example. The problem with interoperability is that programs used to perform analysis will also need to be selective about which properties to read. Geometry is likely important in all models, but many other properties should be left out of models which do not relate to them. Model View Definitions (MVD) are important as they determine what is exchanged and what is not. However, these are determined by trial and error and may not be standard. According to the handbook, there had been 23 efforts as of April 2010 to define MVDs, each for different purposes and programs. Other attempts to allow for interoperability include several XML transports such as OpenGIS and BCF (BIM Collaboration Format), as listed on page 133.
As the interoperability between programs increases, there may be a need to store further information in the BIM model. For example, perhaps it will be possible to notify the mechanical and structural team members if a beam and a duct are losing clearance space in the model before clash detection kicks in. However, as the amount of information being stored in the models increases there may be increased problems in the future with file compression and determining which specific programs need which specific attributes.
EDIT: Comments on Dianna Vogel's post and Bryan Cummings' post.
This comment has been removed by the author.
ReplyDeleteCathlene,
ReplyDeleteThis is a wonderful example on how to summarize interoperability in terms of cross discipline and BIM modeling. It is by no means perfect right now, but the process of converting from AutoCAD to BIM without any major discrepancies may be plausible in a few years. Personally, at my last co-op, I have had to convert drawings from Microstation to AutoCAD which are both drafted in 2-D and still run into problems.
It is brilliant the capacity of BIM to store information such as fire safety as well as structural information so that both a easily accessible in the same program for effective cross discipline work.Personally after seeing what we are already capable of, I am excited to see the transformation of our abilities over the next few years.
Cathlene,
ReplyDeleteNice post. A good read. I liked your comments regarding certain people and professions selecting which relevant information from certain BIM programs to transfer to another program. This could possibly further complicate the already complex field that interoperability is. I guess a possible solution is to implement layers in the transferring and compatibility functions in order for engineers to select what they want to be transferred and forget about the rest.