There are several ways to make time estimates about your project and the most effective and accurate is the Deterministic Method. This method is very accurate because it is based on tasks that you have completed successfully in the past. However you will often have to estimate on work or tasks that you have not performed in the past. Therefore you will have to use the Probabilistic Method. To use this technique, you must develop a work breakdown structure (WBS). Conduct a meeting to develop a list of the pertinent items for the task. Ensure that the necessary skills are represented at the meeting.
Break down all work efforts until costs can be identified. Include labor, materials, support staff, and overhead. Identify incremental costs-costs that would not have been incurred if it weren’t for this project. Identify the major milestones to reach the project goal and include subordinate tasks. Determine what activities must be performed and continue in this manner, from higher to lower levels of detail, until the WBS is complete.
Generally, you need time estimates all project tasks, both routine and non-routine tasks. If you are using a software package, expended and elapsed time estimates will be computed. As a “fallback” plan, for tasks that require a small effort, create an account for wait time for the unavailability of physical resources or team members. The (WBS) must list only the lowest level tasks or sub-tasks that are beneath each major milestone. The results will be rolled up at higher levels by your software.
In this article, we will examine five methods for estimating your projects but your organization or industry may offer different guidelines:
The Delphi Method is used successfully in many organizations. It is a multi-step process that solicits various opinions simultaneously. It is structured to prevent more-assertive and high status team members from dominating the group discussion. Team members work independently to describe completion criteria for a task and make an estimate. These estimates are compiled and the team members are asked to evaluate the task’s completion criteria and to validate the time estimate. These figures are combined and averaged, and the mean time estimate is calculated. At the next meeting, members can revise or justify their earlier estimates in light of new information.
Standards exist in every industry and most organizations have them. Industrial engineers have conducted time-and-motion studies and task analyses to determine the steps involved in doing routine jobs. These analyses can be helpful to estimate similar jobs and tasks. Some organizations estimate that a person assigned to your project for 100 percent of the time will be productive for as much as 75 percent of that time. Others may use lower percentages.
Function-Feature Point Analysis
For software development projects and information system (IS) projects, function point analysis and feature point analysis have added more precision to time estimating. Function point analysis is the most widely used software metric having overtaken lines of code as a standard measure. Estimators classify and count raw function points beginning at certain points in the project after a predetermined amount of design has been completed. Each incidence of inputs such as data entry forms, on-line transactions, optical character readers, mass storage interfaces from other applications is counted. Feature point analysis is an extension of function point analysis that emphasizes complex algorithms that are added to the calculated time needed to write software code.
The PERT Method accounts for uncertainty and time estimates. It uses a probabilistic tool tailored to tasks that have seldom or never been encountered, as in the development of new products. The PERT approach uses the most optimistic time, most likely time and the most pessimistic time. These three estimates are designated to, tm, and tp respectively. For maximum objectivity, gather the three time estimates from three people working independently. The PERT formula explicitly requires three estimates for expended time.
Monte Carlo Method
It is a probabilistic method and integrates the degree of risk associated with each project task to develop a range of possible completion dates. Uncontrollable factors, such as weather and response rates, can be methodically integrated into time estimating. Design a flowchart for an activity or task. The method will randomly approximate the possible outcomes, similar to a coin toss.
For example, let’s say that a task could require three, four, five, six, seven, or eight weeks to complete. Each number has an estimated probability of occurrence. The probability of the task taking three weeks is 10 percent, four weeks is 15 percent, five weeks is 25 percent, six weeks is 20 percent, seven weeks is 15 percent, and eight weeks is also 15 percent. Program the software with these numbers and run some trials. The number that comes up most frequently would be your time estimate. This method is usually reserved for larger projects with which the team has little experience. Software applications and spreadsheets can make these calculations.
Whichever method you use for estimating time, scrutinize each task carefully. If there are uncontrollable aspects of the task or risks of failure, such as when a test is conducted, include a cushion in time estimates. It is always a good idea to break the untried tasks down to a finer granularity than the more familiar tasks and less familiar tasks get more scrutiny.
Please “Like” and share your comments. Additional training resources are located here.