Thank you, it

In the comments to the previous food warmer display article food warmer display "What is silent Gantt chart or why projects are always late" promptly inserted links to videos and slaydkasty describing other reasons causing errors when using Gantt charts. The greatest debate and misunderstanding caused due dates straightening skewed distribution term solution. Under the cut example of the impact on the completion date of the project this very asymmetric distribution. Before we start, let me remind you that in the framework of the examples, the values used for solving the problem of time taken from some previous experience and are objective. food warmer display The examples food warmer display are intended to demonstrate that the drawing in the construction of a Gantt chart only on the expectation food warmer display of a single task completion time, we will get into a trouble. Consider an example in which the most likely time for solving the problem food warmer display is 3 hours. It is quite clear that there is a time that we will solve the problem can not. We need to read the task, see related tasks, food warmer display discuss some issues with the analyst or architect, open IDE, load the project, make changes to the code, sbildit, start, check that everything is working as expected, etc. Suppose, food warmer display in less than 30 minutes, we will not be able to solve the problem. But on the other hand, if you have problems with your computer (unlikely, but still), detection of associated bug, in the absence of on-site personnel with which to align the solution of the problem, food warmer display the term may be increased to 6, and even up to 9 hours. As a result, we obtain the probability density function of the form (along the axis of abscissa time to solve the problem): And this is not a normal distribution. I, for this article, has taken a gamma distribution, there are also others, such as the Maxwell distribution, beta-distribution, etc. Excellent, as I said, the expectation of problem solving time is three hours. Gantt Chart for five consecutive running food warmer display tasks will be: Now back to the gamma distribution and calculations, or (as I did, very much like programming) modeling food warmer display the probability of completion of all phases in a given hour. The first news is joyful, the probability density function assumed a normal distribution, but everything else ... For this to rest understand construct probability food warmer display of completing a given hour: Chance to complete the planned 15 hour is only 20%. To achieve a 90% probability must move completion time is 7 hours (which is 50% of the initial time). And then, in the case of the remaining 10% we will not fit, and a 5% chance we will complete the five tasks after 9 hours. Another sad news is that even the use of the latest food warmer display popular food warmer display methods of three estimates, food warmer display the situation has not straighten. If you do not remember, the main idea of the method consists of three assessments that takes a pessimistic outlook, optimistic outlook and expectation. In this case, the time for completion of the task is determined by the formula: In our case, if we take for the problem as optimistic forecast probability of 30% for the pessimistic - 95%, and for most of the expected 50%, the probability based on the schedule to complete one the problem to a given hour: received: food warmer display Multiplies 3.7 hours, 5 stages and get 18.5 hours. It's already food warmer display much better we fit into this period with 75% probability. A quarter of cases will continue to disrupt timing. Links to materials from the comments to the previous article: 1. Poisson combustion timing 2. Human and mathematical terms straightening Ps For subtraction and tips to become a big thanks to Andrey Potapov, CrazyViper and Cartmendum (last for swearing at my ignorance of probability theory).
Integrated project management food warmer display system for small and medium business August 24, 2011 at 16:33
Thank you, it's interesting. Clearly shows that pure mathematics (method of three evaluations) is not sufficient and the desire to reach the 90% probability food warmer display for the described technique you will likely fly at a reasonable frame (expectations, budget, food warmer display etc.). Made us think and how he is waiting ranks. Vskidku turns out that the estimation accuracy of all people are different, the complexity of the problems are different, it is necessary to take into account the correlation of the complexity and the competence of the performers, the number and nature of the interactions between the participants of the project (chain) and external parties, the motivation of participants and perhaps this is not the whole list. A lot of factors is obtained which is probably working hard to build a mathematical model.
That is why many modern food warmer display methods of trying to move away from the standard planning based on Gantt charts. For example, the control food warmer display method of the critical chain. And the introduction of iterative processes in development, can significantly improve the accuracy of forecasts. If you have missed the first iteration of the 20% is a very high probability, and the entire project you do not meet 20%. This e