Cost Curve Percentages

[unibelle]

I have been given a file that was created before my time and I need help completing these charts. I do not know where this data came from and have done multiple searches with no luck.

Thanks in advance for the help!

[shg]

Those are various spend curves derived from some unknown source and then tabulated for a whole bunch of project durations.

There's a workbook at https://app.box.com/shared/ipkvdyc4bt that implements NASA's Beta polynomial for project loading curves. The functions it contains will give similar results in much simpler fashion.

[unibelle]

I saw this file on a different reply and have been trying to make sense of it. When I edit it to be mid-loaded, the percentages do not add up to what I have in my already populate table - is there a way that they could match up or should I replace the old data with these new percentages?

[shg]

Can you show an example (preferably, post a workbook) of what you tried?

[unibelle]

I set it to mid loaded and changed the number to 4.
This gives me 10.4%, 39.6%, 39.6%, and 10.4%.
My chart has this as 15%, 35%, 35%, and 15%.

[shg]

I added a new tab to the workbook ("Examples", same link) that shows approximate equivalents to your data.

For the bell distribution, I used Solver to minimize the differences between its results and yours:

Row\Col	B	C	D	M	N
3				Beta Bell	Your Bell
4	Periods		**A	0.37076134
5	12		**B	0.25847726
6				Sum	RMS Err
7	Period \ Sum			1.000	0.009
8	1			0.023	0.040
9	2			0.060	0.057
10	3			0.086	0.076
11	4			0.102	0.095
12	5			0.112	0.111
13	6			0.117	0.120
14	7			0.117	0.120
15	8			0.112	0.111
16	9			0.102	0.095
17	10			0.086	0.076
18	11			0.060	0.057
19	12			0.023	0.040

As you can see, it's close. You could apply a similar method for the others.

You won't get exact matches; we have no idea how the other curves were derived. But the Beta model has provenance and is well established.