An interesting project management question…
It’s a hypothetical so bear with me,…You’re building a house with ten parallel tasks that each take on average six months to perform.
So what is the chance that the project will take six months?
Discussion Forum
Discussion Forum
Up Next
Video Shorts
Featured Story
The FHB Podcast team weighs in on Building Science career questions.
Featured Video
Highlights
"I have learned so much thanks to the searchable articles on the FHB website. I can confidently say that I expect to be a life-long subscriber." - M.K.
Replies
0%.
(Actually, if you really could count on those numbers,
0.5 * 0.5 * 0.5 * 0.5 * 0.5 * 0.5 * 0.5 * 0.5 * 0.5 * 0.5, or 0.09765625%. Which is to say a little less than once chance in 1000.)
100% Correct. You get tonight's milkbone.
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
lemme see.... 1 / 1000 = .001
how close is that to zero ?
i protest.... you gave my milkbones to danMike Hussein Smith Rhode Island : Design / Build / Repair / Restore
He was first (in front of you) and he did use math.
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
So why did you ask the question? Since you already knew the answer and all.
Edited 9/16/2009 1:31 am by Mooney
oooh.....oooh....lemmme me..
he wanted to illustrate a point
here's the point....
build like an optimist.... bid like a pessimist
just like your real estate.... the money is made on the buy..... so you can be in position on the sellMike Hussein Smith Rhode Island : Design / Build / Repair / Restore
Mooney — "So why did you ask the question? Since you already knew the answer and all."
To get people to think.
Is there something wrong wtih that?
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
I think it's been legal since January.
As I stood before the gates I realized that I never want to be as certain about anything as were the people who built this place. --Rabbi Sheila Peltz, on her visit to Auschwitz
Nothiing wrong with that . I dont usually ask a question sincerly if I know the answer.
I dont have a problem with it and it doesnt matter if I did.
My opinion is that on forums you have to be clear and precise in delivery like Ive already bragged you are so this just threw me off a bit .
No big deal.
Tim
Don't be blowing Dan's candle out - he gets to light it so seldom!Speaking of rarity, where you been, Stranger?
Welcome to the Taunton University of Knowledge FHB Campus at Breaktime. where ... Excellence is its own reward!
got all kinds of stuff going on... some is even goodMike Hussein Smith Rhode Island : Design / Build / Repair / Restore
When I worked as a Systems jerk in the automotive industry, I would get a good belly laugh numerous times every year. Because in the computer programming fields, it seems that they all believe that humans are capable of having babies in less than a month.It just takes at least 9 women.At least that's how they would manage some of the larger projects -- throw more people in there, maybe they'll get done sooner!
I won't be laughing at the lies when I'm gone,
And I can't question how or when or why when I'm gone;
I can't live proud enough to die when I'm gone,
So I guess I'll have to do it while I'm here. (Phil Ochs)
Great analogy!I will be using that often -I'll be sure to credit the source ;)
I think "the source" is probably long dead.
As I stood before the gates I realized that I never want to be as certain about anything as were the people who built this place. --Rabbi Sheila Peltz, on her visit to Auschwitz
It's a joke in any engineering field. Especially in the programming arena, MANAGERS seem to think that programmers are "plug-compatible" and that one can be substituted for another (or 3 can be substituted for one for 1/3rd the time). Of course, they also believe that the way to handle being behind schedule is to have more meetings.
As I stood before the gates I realized that I never want to be as certain about anything as were the people who built this place. --Rabbi Sheila Peltz, on her visit to Auschwitz
That's assuming each task has a 50% chance of completing on time. Whoever said you want the median not the average (which conventionally means the mean) time to finish was right.Alternatively, if you knew the probability distribution of completion times - specifically, the probability of completing on time for each one, you could plug those numbers in instead of 0.5 for each one.Rebeccah
Whoever said you want the median not the average (which conventionally means the mean) time to finish was right.
Thank youA La Carte Government funding... the real democracy.
> Whoever said you want the median not the average (which conventionally means the mean) time to finish was right.That was me.
As I stood before the gates I realized that I never want to be as certain about anything as were the people who built this place. --Rabbi Sheila Peltz, on her visit to Auschwitz
Throw them all in the ring and let the best sub win. Only the quality will suffer.
Mike
Small wheel turn by the fire and rod, big wheel turn by the grace of god.
Are they all on the critical path? Do any have any slack?
"Put your creed in your deed." Emerson
"When asked if you can do something, tell'em "Why certainly I can", then get busy and find a way to do it." T. Roosevelt
FastEddie - "Are they all on the critical path? Do any have any slack?"
I said the tasks run in parallel. They are performed concurrently. There are no dependencies. That's why I said "It's a hypothetical [case] so bear with me."
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
"average of six months"
some high some low
if it were median it would be 50%
but it aint so
A La Carte Government funding... the real democracy.
0... or real close to zero...
there are mathematical equations to solve this .. but i solve it based on experience... 6 parallel 6-month tasks.... i'd put money against it happening
whatta you think ?
Are you using union labor?
It's actually pretty easy to think about without any math but still most people get it wrong if not really wrong.Think baseball. You have nine players (batters) on a team. What is the chance that each and everyone one of those batters is going to hit for their average on a given night?DanH has hit it right on the head as far as the math is concerned. The chance of that project finishing in six months is a little less than 1 in 1000.
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
Murphy will just visit all of them simultaneously, and you'll finish late.
-d
You need more information than the average time to completion. You need sample data from past projects. Imagine one of your tasks averages 6 months, but on the last four projects that task took 2,9,8, and 5 hours. Compare that to another contractor who's results are 5,7,6,6.
Second guy has a much better probability of hitting the mark.
Averages don't tell the story. Consider the old example: when Bill Gates walks into a room full of people the average income in the room goes way up, but that says nothing about the distribution.
mike585 - "You need more information than the average time to completion. You need sample data from past projects. Imagine one of your tasks averages 6 months, but on the last four projects that task took 2,9,8, and 5 hours. Compare that to another contractor who's results are 5,7,6,6."
Not really Mike, the hypothetical data from past jobs gave us the average completion time of 6 months for each of the hypothetical tasks. That's all we need to answer the question as to what are the chances of it completing in six months.
And beyond this case there is a huge problem with using only the most recent data points to predict probability or a trend. A general rule in statistical analysis rule of thumb that you "never take a trend from the end of a polynomial". Also using data in such a fashion is also what W. Edwards Deming has taught us is tampering and the problem tampering leads to he illustrated with his Funnel Experiment.
"Averages don't tell the story. Consider the old example: when Bill Gates walks into a room full of people the average income in the room goes way up, but that says nothing about the distribution."
On this we are in agreement. In fact I bring this up because of a book I just read you might be interested in: Sam L. Savage's The Flaw of Averages: Why We Underestimate Risk in the Face of UncertaintyView Image.
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
I'd agree with you if only you were right. You are talking about PERT. You can't do PERT without knowing the standard deviation of your task completion data.
mike585 — "I'd agree with you if only you were right. You are talking about PERT. You can't do PERT without knowing the standard deviation of your task completion data."
No I was not talking about PERT. PERT is a step or tow beyond what the question I asked was about. You don't need PERT to answer the question the way I asked it and DanH understood it and hit it right on the button.
Had I given the standard deviations for the resource tasks and asked 'when will the project most likely finish' then we could possibly be talking about PERT.
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
Of course, "average" needs to be more precisely defined. The value of interest is the "median", not the "mean".
As I stood before the gates I realized that I never want to be as certain about anything as were the people who built this place. --Rabbi Sheila Peltz, on her visit to Auschwitz
Is this a practical problem?
0% chance
or a logic and statistical rhetorical Q?
Welcome to the
Taunton University of Knowledge FHB Campus at Breaktime.
where ...
Excellence is its own reward!
From: Piffin — "Is this a practical problem?
0% chance
or a logic and statistical rhetorical Q?"
A project with 10 concurrent tasks all having a average completion time of six months? How realistic is that?
But then again it not at all unrealistic to think that we might have a real life project with 3 concurrent tasks that have an average completion time of 6 months. That could possibly happen. What the chance then of the project completing in six months?
12.5%
How about to two concurrent tasks?
25%
So while yes it is rhetorical in that it's intended to illustrate a point and get us to think it does have more realistic and real world implications.
-(and yes in deference to mike585 if we wanted to make actual predictions about when the project would most likely would be finished we would need more information, the standard deviations of the resource tasks, to perform our calculations.)-
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
slim to none
the hypothetical question shouldn't be if it can be done. It should how will it take to complete the job with all the parallel tasks.
That is what we are being paid to do.
You are starting to sound like some of my past clients
..."but we thought you can get it done by Thanksgiving"
Ummm..Jerald....does my "slim to none" response get me a milkbone?
jimAKAblue — "Ummm..Jerald....does my "slim to none" response get me a milkbone?"
Well yeah I guess I can give you a "virtual" milk bone for being correct. But you're late with your answer. That kind of procrastination is one of the reasons why these tasks finish in longer than average times. You're holding up this whole project!
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
"You're holding up this whole project! "
I hate it when I do that .
JJH — "You're holding up this whole project! "
The funny thing, or not so funny thing is doing stair and railings like we do we are often in a position on a lot of the projects we work where we are right there smack dab on critical chain towards the tail end of the project executing a task that has to be completed before a whole bunch of other finishing tasks on the project can be started so I know the feeling pretty damn well.
(The other task we sometimes trade spots with on the critical chain is the floor finishers.)
(...and if anyone out there that wants to correct me saying the correct term is "Critical Path" not "Critical Chain" yeah I know what you mean but I very deliberately meant and used Critical Chain. Do you know what I mean?)
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
Edited 9/16/2009 8:47 pm ET by JerraldHayes
Didn't you mean to type "dilbertly"?
As I stood before the gates I realized that I never want to be as certain about anything as were the people who built this place. --Rabbi Sheila Peltz, on her visit to Auschwitz
DanH — "Didn't you mean to type "dilbertly"?"
Well not quite but that certainly would have been apropos if I was witty enough to think that up in the first place. And that wasn't my only spelling error in that post either . My GF had just pulled in the driveway so I posted without spell checking. Back from dinner I just fixed it. Funny thing to is she's a certified PMP so guess what we were talking about at dinner?
View Image
The Woodshed Tavern Backroom
The Topics Too Hot For Taunton's Breaktime Forum Tavern
(Message redacted before transmission.)
As I stood before the gates I realized that I never want to be as certain about anything as were the people who built this place. --Rabbi Sheila Peltz, on her visit to Auschwitz
My arms get tired and my back aches.
As I stood before the gates I realized that I never want to be as certain about anything as were the people who built this place. --Rabbi Sheila Peltz, on her visit to Auschwitz
No math needed. Common sense and experience is all it takes. Using math here is similar to measuring, marking, cutting trim. I don't know the lengths of the window casing. I hold them up and mark them in place. Don't know the lengths, don't need to know the lengths, don't care that the exact lengths are. No need for measuring, or math.
All I know if they fit perfect. And the chances of that job working out are zero. Vic