Agile Estimation: An estimate of any work to be done, can be accurately guessed if it is
compared to someone’s else work completion duration. Agile estimation basically says that
most of our estimation is dependent on others. So we are a bit relative estimates. So the
story time defines the duration of any work to be done. It can be one hour, a day or even a
week. While story value defines the value or worth of the project. In order to proportionate
the projects we give them the story points. Fibonacci numbers are the sequence of the
number in which sum of two numbers give the next upcoming number. This sequence is
quite similar with human beings estimation. [ CITATION Ras14 \l 2057 ]
In order to give deliver value early, we must have to consider both some measure of value
and some measure of time as well. The story points and value are put in the backlog of the
scale, the second story is very important and is always put this on any end of the scale. So
point story tells about time to take to do the story and value point tells the importance of
the story delivery. Some time, the high value story takes a long time. To overcome this
problem, the bang of the book score tells us about the way to give the most value story in
very shortest time. Iterations are done to estimate the time, value and cost as well. The
value is decreased as the time passes, which increase the expenses on the project. This thing
helps customers to know that further working on the ongoing project will be beneficial or
not and the time comes when value drops to zero and they eventually finish the project.
[ CITATION Muh17 \l 2057 ]
Questions:
1. What should be firstly estimated between absolute metrics and relative estimates?
2. For the first iteration, how is velocity estimation done?
Responses:
1. Mostly, the workers prefer to do the relative estimation for their success as it more
quicker than the absolute estimation. There is a benefit of relative estimation
because the workers remain focussed. Simple example can be given as if a group of
friends visit the city where there are a lot of sky scrapers, they start guessing the
height of one of the building. There are high chances that they don’t get it right at
first and may become dispersed in their ideas as well. But if they are told with the
height of another building which is quite near to that building then they will try to
help each other in judging the more accurate height of that building. So it’s not just
about the effort or its duration , uncertainty and complex combination. All things
should have to be taken into consideration while estimating. [ CITATION Muh15 \l 2057
]
2. According to the general guidelines, initial velocity should be estimated as one third
of the time that is available during the iteration. Time that is spent in meetings,
design, research, documentation, collaboration and rework should be considered
into that time which is allocated to a User Story. For example, if six programmers are
supposed to work I a sprint of three weeks, then the total working days will become
15 days excluding the weekends. So total time for the iterations would become 90
days if we multiply 6 into 15. This means that 30 days time should be allocated for
first iteration (90 x 1/3=30). This is done for the sufficient buffer for project overhead
and estimation inaccuracy if it is required later on. If this is underestimated, then it
will result in increase in the velocity after first iteration because of the addition of
new features. But if it is overestimated then there will be decrease in the velocity
due to removal of the features. In this case, the Scrum team will use the first
iteration velocity as the reference. [ CITATION Ala14 \l 2057 ]
Reference:
Hamouda, A. E. (2014). Using Agile Story Points as an Estimation Technique in CMMI
Organizations. 2014 Agile Conference, 16-23.
Muhammad Usman, E. M. (2015). Effort estimation in agile software development: a survey
on the state of the practice. EASE '15: Proceedings of the 19th International Conference on
Evaluation and Assessment in Software Engineering, 1-10.
Muhammad Usman, J. B. (2017). International Journal of Software Engineering and
Knowledge Engineering. An Effort Estimation Taxonomy for Agile Software Development,
641-674.
Popli, R., & Chauhan, N. (2014). Agile estimation using people and project related factors.
2014 International Conference on Computing for Sustainable Global Development
(INDIACom), 564-569.
