Every Project is a Gamble: Know the Odds Before Placing Your Bet
Mike Ross · 2007
(Abstract) This paper starts with the premise, The only thing certain about project per- formance is uncertainty. Over 25 years of software industry experience and project data tells us that cost and schedule estimating is essential to diligent affordability management and project management. It also tells us that projects behave according to certain dynamic properties; that duration, effort, cost, and defects are all inexorably linked (correlated); that these correlations can be expressed as functions of product and project attributes; and that, prior to project completion, everything is uncertain. A primary goal of any thorough project estimating process should therefore be to not only yield estimated values for these metrics; it should also indicate whether or not these estimated values satisfy their corresponding pro- ject goals within some corresponding specified confidence limits (probabilities of success). This paper describes a new chart (we call it a Ross chart™) that shows the confidence limits (probabilities of success) and goals satisfaction of two correlated random variables. Using an effort vs. duration tradeoff relationship example, the paper describes a method for stochastic (probabilistic) treatment of bivariate relationships. It then presents an example project esti- mating scenario using a system of three synchronized Ross charts to show the degree of risk associated with a particular estimating solution. In other words, this paper presents a proc- ess (methods and tools) for performing confidence-driven estimating. I. Nomenclature E α = effort exponent t α = duration exponent () , ab δ = specific defect vulnerability during the time interval beginning at a and ending at b γ = gamma (economy exponent) η = specific efficiency η = specific efficiency expressed as a random variable η = specific efficiency expressed as a vector of appropriately distributed possible outcomes E ϕ = defect effort exponent t ϕ = defect duration exponent () , ab Φ = total count of defects that occur during the time interval beginning at a and ending at b ψ = size/efficiency ratio