“Not too shabby.”

(excerpted from The PipelineThe Pipeline: A Picture of Homebuilding Production is available on the publisher website (virtualbookworm.com), and the author website (thepipelinebook.com), as well as amazon.com and barnesandnoble.com)

 

The intrepid, results-based consultant reached across her notebook computer and recovered her programmable HP-12C – the most recent version of the weapon she had long carried – from the superintendent who had been fiddling with it.

“Figure it out yet?”

“It’s an HP-12C”, said the superintendent.  “It uses RPN.  It’s been around for more than 25 years, almost as long as me.  It’s definitely old school;  it might be a coelacanth.”

“Definitely old school, but it’s not a coelacanth”, she said.  “That’s because it still works better than anything else that has come along.  Unlike the algorithms still in use by most scheduling programs.  There is a difference between a classic and a fossil.

“But – to think that I thought you guys only counted on your fingers and toes.  So, where were we?  Ahh, yes.

“As I recall,  RB Builders takes close to 180 days to complete a house, despite a job schedule that is intended to be only 120 days.  Since 120 days reflects the sum of highly-confident task durations, it stands to reason that about half the time, those jobs could be completed in around 74 days.  As you remember, we characterized the 180 day actual cycle time as a picture of all the delays cascading through the overall job schedule.

“What if we reduced the job schedule from 120 days to 74 days – a difference of 46 days – and then extended the new, shorter job schedule to 97 days, by inserting a 23 day project buffer at the end of the job schedule?

“What would happen?”

“It would save 21 days out of the schedule”, said the superintendent.  “Didn’t need a calculator to figure that one.”

The CFO looked at the intrepid, results-based consultant.  “Our actual cycle time is 180 days, not the 120 days called for by the job schedule.  Are we to draw from this that taking half the padded duration out would reduce our actual building time from 180 days to 157 days?”

“I don’t think that’s what she’s saying”, said the senior superintendent.

“What she’s saying”, said the VP of Construction, “is that we can reduce our actual cycle time by an average of 83 days, from 180 days down to 97 days, and in the process, make the schedules far more reliable, by cutting out a lot of the hidden safety that the schedule consumes anyway, because of the human behavioral tendencies we talked about before.  Remember, the only reason we experience 180 day cycle times on jobs with 120 day schedules is variation and uncertainty;  eliminate the negative effect of variation and uncertainty, and you can do exactly what she is saying.”

“Save 21 days on the schedule, cut 83 days out of the actual build time”, said the senior superintendent.  “Not too shabby.”

The intrepid, results-based consultant smiled.

“That’s right”, she said.  “Without doing anything else differently – with the possible exception of making all of your jobs easier – you would cut your cycle time almost in half on every house you build.  Without adding one iota of production capacity – from either a resource standpoint or a financial standpoint – productivity would almost double.  Which means, the rate of physical throughput – the rate of closings – would likewise almost double.

“In today’s real estate market, we don’t know what that means for Gross Income or Net Income, because we don’t know what kind of margins it would take to acquire all of the sales necessary to fully utilize the additional capacity that would be the result of that much increase in productivity.

“What we do know is that – as long as the Gross Income Margins are positive – all of the Gross Income above breakeven would drop straight to RB Builders’ bottom-line.

“Not too shabby.”