20.8 Comparison of Results from Different Methods


20.8 Comparison of Results from Different Methods

Here are the five schedule estimation methods we've used in this chapter:

  • Basic Schedule Equation

  • The Informal Comparison to Past Projects formula

  • Jones's First-Order Estimation Practice

  • Software tool calibrated with industry-average data

  • Software tool calibrated with historical data

Figure 20-4 shows graphically how these different schedule estimates compare.

image from book
Figure 20-4: Ranges of schedule estimates produced by the methods discussed in this chapter. The relative dot sizes and line thicknesses represent the weights I would give each of these estimates. Looking at all the estimates, including those that aren't well founded, hides the real convergence among these estimates.

At first glance, the schedule estimates in this example don't seem to converge very well. One refinement that could be made is based on the Basic Schedule Equation used to generate the top line. The total range shown in Figure 20-4 is for coefficients in the Basic Schedule Equation that range from 2.0 to 4.0. If you review your historical data, you can assess how much that coefficient should actually vary. The past projects' coefficients in this chapter's example have ranged only from 2.7 to 3.7, which narrows the schedule range produced by the Basic Schedule Equation to 11.6 to 14.1 months.

In this case, I would again heavily weight the estimate produced with Construx Estimate both because it is an estimation-science approach and, more important, because it's based on historical data. I would weight the Basic Schedule Equation and the Informal Comparison to Past Projects formulas second. I would not use Jones's First-Order Estimation Practice approach at all for a situation in which I have better data.

Figure 20-5 shows the convergence of the schedule estimates when we remove the overly generic data.

image from book
Figure 20-5: Ranges of schedule estimates produced by the most accurate methods. Once the estimates produced by overly generic methods are eliminated, the convergence of the estimates becomes clear.

Based on this convergence, I would present a range of 11.5 to 14 months, and I would probably not provide an Expected Case within that range. The schedule techniques in this chapter are all early-in-the-Cone-of-Uncertainty techniques, so not providing a single-point number would normally be acceptable at this early stage of a project.

Tip #97 

Remove the results of overly generic estimation techniques from your data set before you look for convergence or spread among your estimates.




Software Estimation. Demystifying the Black Art
Software Estimation: Demystifying the Black Art (Best Practices (Microsoft))
ISBN: 0735605351
EAN: 2147483647
Year: 2004
Pages: 212

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net