Size Doesn’t Matter

Hello BIMfans,
As you know, I am currently using the Asset Information Model I produced during this blog to manage my home.  So far my information model has been used to support day-to-day operation, minor works such as getting my gutters and fascia replaced, as well as product registration and replacement, in line with my  Model Purposes and Data Requirements.  But has it been worth it?  Well, let’s try and work it out (don’t worry, I’ll do the sums).

ROI_Cover

Well, first thing’s first, how much did my information model cost? Well, because I did it myself, there isn’t really a figure I can apply.  However, as it has taken a while to produce, instead, I decided to use a different form of currency; minutes instead of pounds (after all, time is money!).

Quite often when talking about the blog I am asked “How long did it take to model all that!?”, and while it has taken over a year to produce, it was done in my spare time, with a lot of trial and error.  If I lost all of this information (please, please, NEVER let this happen!) I believe I could reproduce everything within four working days (1800 minutes).  This aligns with a New Zealand case study that took three days to model and write-up social houses.  So the real question is:

How long will it take for my information model to save me 1800 minutes?

My plan has always been for my information model to be useful, so I am confident that there is a return on investment; I just need to work out how long it is.  When I wrote my Model Purposes and Data Requirements, I decided that I would use my information model for the registration, operation, maintenance, repair, and replacements.  So let’s see how it has helped in those areas.

Registration

While trying to write this post, I discovered that I have registered a surprising number of components.  Excluding loose furniture, I have registered 104 (yes, 104) manageable components and assessed their condition.

Traditionally to keep track of the condition of these components, frequent property condition surveys would have been needed.  While I can find examples from several housing associations who undertake annual condition surveys, I have decided to instead follow Cardiff Council‘s own Asset Management Plan; using their conservative 2-3 year programme.  Having done surveying in a past life, I’m confident that to survey and capture all the relevant COBie information for these 104 components would take a day and a half (average of 6 minutes per component), whereas updating my information model would only take me half a day.

COBie
Having this information as the basis really speeds up the surveying of my home.

This results in a saving of 450 minutes every 2.5 years, or an annual saving of 180 minutes. (This alone would provide a return on investment within 10 years)

Operation, Maintenance, and Repair

Working for BRE, I’m sure you won’t be surprised to hear that I want to reduce my home’s carbon footprint.  A goal of mine is to improve my home’s EPC score from a D (65) to a B (86).  To do so, I am going to need some tradesmen every now again (but never again from Rated People, grrrrr).

My information model has been great for solving operational problems.  For example, when I had my gutters and fascia replaced, the quoting tradesmen raised an issue.  I live in a thin terrace house, with no external access to the garden.  After quoting for the work, they each wanted to re-visit and check whether their six metre scaffold pipes could get through the house.  I didn’t want to book another half-day off work, so I instead used my information model to demonstrate that it was possible.

Once I proved this, I was happy to accept accountability that the scaffold would fit, and saved myself from having to book a half-day off work.

If I assume I’d get tradesmen in every two years, this results in a saving of 225 minutes every two years or an annual saving of 112 minutes. (This alone would provide a return on investment within 17 years)

Replacement

As I’ve mentioned, I have 104 manageable components, each of which will ultimately fail and need to be replaced (What Asset Managers would call a ‘trigger-related event’).  So I have done some research and calculated my components’ service lives.  For example, my two extract fans, have a pitiful service life of 10 years, while my internal doors have an impressive 100+ years.  Whenever I need to replace any of these components, I will be saving time.

Picture this scenario, while I am out shopping I suddenly receive a call.  My wife has rung to say that on her way out she noticed a bulb had blown in the living room, and asked if I could buy a new one on my way home.  There is just one problem, I cannot remember whether my living room pendant takes an Edison (E27) or a screw fix (E14) bulb.  So I have two options:

  • Option 1:  I drive home (30 minutes), grab a chair and check the bulb (2 minutes, working from height), drive to B&Q (30 minutes, other hardware stores are available), buy the new bulb (5 minutes), drive home (30 minutes), and then install the new bulb (2 minutes, working from height) for a total of 99 minutes.
  • Option 2: I check my information model on my phone and confirm the bulb type (2 minutes), buy the new bulb (5 minutes), drive home via B&Q (30 minutes, other hardware stores are available), and then install the new bulb (2 minutes, working from height) for a total of 39 minutes.
The Value of Asset Capture
Oh look, my pendant has an E14 to E27 adapter on it, that means I could have bought either!

So by referring to my information model each time I replace a component and purchasing replacement on my way home, I am saving myself 60 minutes and halving the amount of time I am working from height.  While an hour might seem excessive, this is quite conservative compared to the case study on the BIM Task Group Website from Manchester City Council, where the Bulb Replacement Case Study reports a saving of 8 hours (480 minutes) per replacement.

replacement

Based on available service life figures I found online, I produced this table which groups components by their IFC type, and shows an annual saving of 205 minutes. (This alone would provide a return on investment in under 9 years).

Results

Having looked at my various Model Purposes, I have come up with the following annual return on investment thanks to the efficiencies I have gained through using my information model.

Registration = 180 minutes
Operation = 112 minutes
Replacement = 205 minutes
Total = 497 minutes

Which, when compared to a production time of 1800 minutes, gives a return on investment in 3.62 years.

And there we have it, by using conservative figures a reasonable return on investment could be calculated.  When you consider that this excludes all loose furniture, and does not factor in any accidental damage, my information model will have saved more time than that it took to produce in little over three and a half years; Fantastic!

Note:  If you have any comments regarding other efficiencies I could factor in, or the figures I have used to calculate my ROI, then please let me know either on Twitter, or by commenting below.

Advertisements

PLQ 3.2 – Classical Conditioning

Hello BIMfans,
When I made my information model I didn’t want it to be just for show, it wanted it to be a (useful) tool to manage my home. That is why I was very pedantic fussy particular about what information I needed by forming several Plain Language Questions, my Model Purposes, and my Data Requirements. Since this blog’s outset, one clear output I had in mind was to use the information model to manage any repair or replacement work needed within my home. To do so, I will need to form a preventative maintenance schedule; time for some Classical Conditioning!

I’m no Pavlov but I am known to drool over good information.

When I first wrote my Data Requirements, I was keen to incorporate a way to capture the condition of each of my components. The problem was, I needed a way to record this consistently; luckily for me, there is a way to do this.  BS 1192-4, the British Standard for COBie, includes some additional attributes under table 14, which are also included as part of IFC4 Schema, under the Pset_Condition property set.

Pset_Condition
AssessmentDate, when the assessment was completed YYYY-MM-DD;
AssessmentDescription, qualiative description of the assessment; and
AssessmentCondition, the condition:  Very Poor, Poor, Adequate, Good, or AsNew.

So, a plan was formed.  When I produced my components, each of these attributes were added to the one I intended to manage. As a result, this information appears in each of my graphical models, IFC exports, and COBie files (the joy of a single source of truth).  As the majority of these components were assessed when Chris John undertook a (very thorough) property condition survey before we bought the home, there isn’t much additional information to be collected.  The only exception being new items such as my Nest thermostat and Philips Hue bulbs that have been installed since.

As you can see, this window was surveyed August 2015, as part of the property condition survey, and it’s in a pretty good condition.

Using these assessment attributes, I can manage each of these components and develop my preventative maintenance schedule. For example, using the AssessmentCondition attribute, I can filter and identify any Very Poor or Poor components. Of the 100+ manageable components I have in my home, I can use Revit‘s scheduling function to filter this information down to just those components and form a manageable schedule.

NOTE: I could have done this using my COBie file, but I won’t.  COBie isn’t a data management tool. Until I acquire an asset management system, using the information embedded in my graphical models, as I have done, is the best solution.

By federating my models, I can create a single schedule showing all of the Very Poor and Poor components in my home.

And there we have it.  By using the information that I have already populated within my information model I was able to create a preventative maintenance schedule highlighting what components need to be repaired or replaced. This means that PLQ3.2 is complete; Woohoo!

Operation and Maintenance
3.1 What are the sizes and condition of the windows & doors?
3.2 What assets are in a poor condition?
3.3 What costs can be attributed to my assets?
3.4 What are the most cost effective thermal improvements that could be undertaken?

Now that I what needs doing around my home, I wonder how much it’ll cost to fix…

PLQ 3.1 – Outstanding Openings

Hello BIMfans,
After completing my BIM Level 2 deliverables and publishing my Project Information Model it’s about time I put my model to work.  I have now entered the operational stage, which means I have some new Plain Language Questions to answer:

Operation and Maintenance
3.1 What are the sizes and condition of the windows & doors?
3.2 What assets are in a poor condition?
3.3 What costs can be attributed to my assets?
3.4 What are the most cost effective thermal improvements that could be undertaken?

So without further ado, let’s tackle Plain Language Question 3.1 “What are the sizes and condition of the windows & doors?”

construction-architecture-fails-mistakes-7
This is what happens when you accidentally flip your door vertically instead of horizontally in the model…

Now lucky for me, I have (technically) already answered this question as my Architectural COBie file has all of this information I need already included.  However, I don’t feel that COBie is the best way to communicating this information as I haven’t imported it into a suitable asset management tool (yet) and the information needed is on several different sheets.

doorwindowcobie
Sizes are on the type worksheet, while assessment information the attribute worksheet.

So I have instead decided to answer this question with a schedule.  As I have all the information within my model, it is just a matter of deciding how I want to structure my schedule and what information would be useful to include within it.

To answer my Plain Language Question, I will need Size (Height and Width) as well as Assessment Condition as a minimum.  However, to relate these back I will also need a reference for each component (IfcName), and a type reference (TypeName).

In addition, to make the schedule useful it is worth providing the type description (IfcDescription) as well as indicating which space each component is in, as well as the other assessment details (AssessmentDate and AssessmentDescription) to provide sufficient content.  This leaves me with a schedule that looks like this:

DoorSchedule Header

This scheduling format provides me with enough information to assess each of my doors and windows, find their location within my home, and includes enough information to arrange for a replacement if required.  Luckily for me, I could put all of this information onto a single A3 sheet, which can be accessed here.

SheduleDrawing.JPG
*Dan’s perfectionism sense is tingling* Why couldn’t I have only 4 door types??

 

Note:  You’ll notice that I have named this deliverable: 7001-BBH-ZZ-ZZ-DR-A-6001. While full of schedule information, it is a drawing.  If I had exported this information as a spreadsheet then it would have been a schedule, using the SH file type. However, as the schedule’s information cannot be referenced (can’t import this into Excel) and it is placed on a title block, it’s a drawing.

From this schedule, it is plain to see what my back door (DoorType04, Door05) and my bathroom window (WindowType08, Window09) require work.  There is visible rot on my back door and something that I had never noticed before about my bathroom window is that instead of a sill the installer used is actually a skirting board!

WindowSill.jpg
What kind of monster would do this??

And there we have it.  By using the information that I have already populated within my information model I was able to create a Door & Window Schedule with all of the information to answer this Plain Language Question and picked up which of these components need work. This means that PLQ3.1 is complete; Woohoo!

Operation and Maintenance
3.1 What are the sizes and condition of the windows & doors?
3.2 What assets are in a poor condition?
3.3 What costs can be attributed to my assets?
3.4 What are the most cost effective thermal improvements that could be undertaken?

Now that I have used by schedule to assess my doors and windows, it’s about time I look at what else needs doing around my home…

Update: Removed ‘OK’ for ‘Adequate’ under Assessment Condition to suit BS1192-4 permitted values as opposed to the IFC4 documentation example.  Thanks Nick!

Update: Turns out I had mislabelled by window openings showing them bottom hung as opposed to top hung (Ooops!); a quick check of BS8541-2 and it is all fixed.  Thanks Chris!

Update:  Changed ‘Mark’ to ‘IfcName to improve information consistency in all of my deliverables.  Tags, Schedules, and COBie now use IfcName consistently.