BIM: Less is More

When it comes to BIM, Less is More.

By that, I mean that it is possible to have too much information. BIM often ends up being way too complicated. As I see it:

  • quality beats quantity
  • simplicity beats complexity
  • economy beats excess

Sharp prose or oration rely upon a minimum of words, but they have to be the right words. The same applies to BIM. Unfortunately, BIM often becomes the equivalent of a long, rambling story where the important parts of the narrative are lost.

Without getting too philosophical, I like the the approach of Colin Chapman (Lotus cars) of ‘simplify, then add lightness’ to make faster cars rather than adding power and weight, and believe the same can apply to BIM (and technology in general)

Caterham Lotus Super 7 Roadsport

photo credit: Caterham Lotus Super 7 Roadsport via photopin (license)

Client Deliverables: Everything including the kitchen sink

As I explain in my posts ‘The rise of the BIM consultant‘ and ‘BIM Befuddlement: LOD vs LOD‘ a natural tendency is to assume more data is better.

In fact, the BIM data that most building owners want for operational purposes is really quite simple. All the nuts & bolts that went into the construction are irrelevant.

Everybody does not need everything

Within a design & construction context, some model data is common & necessary for everyone, and some data is only relevant for very specific purposes or for a few people. Therefore, it makes sense to structure the models so that the common and essential data is lean, well disciplined and easy to share/manage, and the more specific data is separate and can be overlaid or loaded on demand.

For example: on a building project, the existence & position/extent of a steel beam is relevant to everyone, including architects, engineers, contractor & subcontractors.
But the exact fabrication details of weld type, fillet radius, fasteners, slotted holes, cleats and so on are primarily relevant to one party i.e the steel fabricator and to the design engineer for checking the structural integrity. Everyone else doesn’t really care.

Why is it so ?

Nature of people who work with BIM

BIM people tend to be naturally technical & detail-focused, and sometimes cannot see the forest for the trees. In other words, very technical or detailed issues can take on  greater importance than fundamental or broader issues.

Because you can…?

Often it comes down to the thinking that because you can model to the nth degree, then you should. For example, I recently saw an architectural model of an entire hospital containing everything down to the individual keys on a computer keyboard and the curly cable, repeated on each desk in the building. As a result, the model was virtually unusable as it took so long to open & regenerate.

The lazy way out

It can be easier to import a very detailed model from an object library, manufacturer’s website, or copy/paste it from another project. It can be quicker to take an ‘all or nothing” approach, even if everything is not necessary.

Consultant fees

It can be in the interest of design consultant or BIM consultants to over-emphasise the complexity of BIM as a means to generate or raise fees. In this case, the BIM specification can be inflated and padded out to include excessive amounts of modelling or data deliverables, for which the client will have no conceivable need or receive no benefit.

Big deal? Where’s the problem ?

There are a number of issues in having too much information (particularly geometry) in a model:

  • too much data overwhelms downstream users
  • important issues are obscured by superfluous information
  • computer performance
  • too much detail too early suggests the design is fully resolved
  • represents wasted effort/time
  • brings in unwanted clutter & increases model management overhead
  • takes longer to make design changes
  • BIM is only accessible by few people

Overwhelming amounts of data

Superfluous information obscures more important information and overwhelms end users. This applies particularly to building owners and the transfer of data into asset management systems, but also to others involved in design & construction.

As I’ve said- this information that building owners need is quite simple, and a well disciplined model that contains only relevant information is much more important than every single detail.

Computer performance

Whilst ever-increasing computing power means that performance of BIM applications with very large models becomes less of an issue over time,  Parkinson’s Law applies i.e the amount of data grows to fill the available space (or processing capacity); but it is important to make sure that models are well within performance constraints.

Facilities management applications are designed to work with much simpler (and smaller) data sets than is typically produced in design & construction BIM, so the performance of these systems can suffer with too much data.

Complexity of Process

Too much data tends to complicate underlying or dependent processes. The challenge is to enable reliable and useful information to be quickly located, and to make it relevant to the end user.

Confusion

If only a few BIM ‘experts’ can understand the model complexity, or find the data that is required, then there is unlikely to be a wide benefit in using BIM.

Model clutter

Most BIM applications are susceptible to accumulation of clutter, which slows down, distracts or confuses people working on the model.

Revit aficionados will know that some weird object categories snuck into the Autodesk templates- some might still be there. For example, there were object categories not just for generic cars, not even for particular types of cars, but specifically for Dodge Viper cars, such as chassis, tyres, reflectors etc.

Maybe not the end of the world, but the countless people working from these templates had to scroll past, ignore or select/deselect these categories, and therefore it is a bit more than an annoyance and wastes time.

dodge-viper-object-categories-400px1

BIM Strategy Tips

  • What is left out of the model is as important as what is included.
  • Question what must be modelled for each project & focus on what really matters
  • There is nothing wrong with detailing or annotations- not everything has to be modelled.
  • Keep models lean, disciplined & proactively maintained
  • Structure models to allow selective loading of data
  • Model geometry where only essential, otherwise use object attributes.
  • Use placeholders with links to more detailed data if it is relevant to just a few people
  • Do not be overambitious i.e that the model will contain everything
  • Define a realistic modelling scope & aim for highest quality
    (rather than trying to do everything and none is good quality)
  • If you are a manager, business owner or otherwise have people working for you using BIM, do not assume that everything being modelled is essential.
    Gain basic skills so that you can review a model and question what is being done & why.