Balancing Multiple Perspectives (BMP)

SEMQ is going to move to www.semq.eu: save the URL!

�Tracking & Control� activities in software projects are most often based just on two dimensions of analysis: time and cost, excluding other ones such as quality, risks or the impact on society, taking into account the stakeholders' viewpoint in a broader sense, as expressed in Performance Management models such as EFQM or the Malcolm Baldridge models.
The identification and selection of the required number of viewpoints for representing a project more adequately is an issue which needs to be addressed in planning a measurement program: �Did we plan and gather data from an appropriate number of indicators?�. A paper discussing these aspects in detail is this one, using as an analogy the Egyptian paintings. Knowing that its source was a 3D figure, even a casual observer is aware that something is missing in the painting, that is, the depth of the image.
Project management, whatever its application domain, should report on several perspectives, since the use of only two dimensions (usually time and cost) represents an overly simplified view of a much more complex reality. To concurrently handle multiple project dimensions (or perspectives) including, for instance, quality and risk, a multidimensional project management approach is needed. Our proposal for realizing such multidimensional approach is given by the QEST/LIME models family.
Because of budget constraints, the key improvement point should be not reducing tracking & control costs (it would only reduced also the probability for the project to achieve its profitability goal) but optimising that amount of budget balancing those control mechanisms across several dimensions of analysis.

BMP (Balancing Multiple Perspectives) is a procedure with the aim to help project managers to choose a set of project indicators from several concurrent viewpoints. The four steps to perform are:
1) Determine the dimensions of interest in the project: at least three dimensions - four or five - would be a good idea, such as in EFQM, Baldrige, BSC;
2) Determine the list of the most representative measures associated with each dimension;
3) For each of the measures selected, identify which other control variables might be impacted negatively (e.g. counter-productive impacts: for instance, higher quality will often mean a greater initial cost or longer project duration; the same applies to cost and risk;
4) Figure out the best combination of indicators and the causal relations between them in order to build a measurement plan for the project.

This is an example of a generic BMP template taking into account 4 possible dimensions of analysis, with the positive (), neutral (=) or negative () impacts:

and this is an example of instantiation of BMP - having chosen Time (T), Cost (C), Quality (Q) and Risk (R) as the four perspectives - with possible causal relationships from a selected list of metrics (just 13 out of an initial list of 67, see PSM Guide v4.0b):

Benefits applying BMP
Main benefits from the application of BMP can be:
enlarge the "tracking & control" usual scope to more than the usual two perspectives (Time and Cost), having more inputs for cause-effects analysis
have a starting point for building your own "strategic map", as in the BSC logic
put in place a more granular view on the ROI from "tracking & control" mechanisms and introduce/maintain a balanced number of controls on your processes, changing the said "the more you measure, the more you control" into "the better you measure, the more you control".

Experimentation/Trial
We are conducting trials with BMP: we have prepared a questionnaire and a basic set of 14 analysis indicators for having the main goals to focus on:
the # of perspectives and metrics (chosen and desired) and how they could be influenced by respondents' role
the analysis of causal relationships among processes, using metrics in a Balanced Scorecard logic
the costs for �tracking & control� process within the organization
the Questionnaire is available in several languages (currently in English, Italian, German, French, Turkish) in order to facilitate the comprehension by interested people

Those interested in applying it to their organization can contact us for receiving assistance simply in change of the returned filled questionnaires (yet anonymized) for increasing our database for further analysis and approfoundishments on this issue, "redistributed" through technical reports and publications.

Publications:
L.Buglione & A.Abran, Multidimensional Project Management Tracking & Control - Related Measurement Issues, Proceedings of SMEF 2005, Software Measurement European Forum, 16-18 March 2005, Rome (Italy), pp. 205-214 click to download the paper Click to read the abstract
L.Buglione & A.Abran, Improving Measurement Plans from multiple dimensions: Exercising with Balancing Multiple Dimensions - BMP, 1st Workshop on "Methods for Learning Metrics", METRICS 2005, 11th IEEE International Software Metrics Symposium, 19-22 September 2005, Como(Italy) click to download the paper Click to read the abstract
R.Dumke, A.Abran & L.Buglione, Suggestions for Improving Measurement Plans: First Results from a a BMP Application, Proceedings of SMEF 2006, 3rd Software Measurement European Forum, 10-12 May 2006, Rome (Italy), pp. 209-224 click to download the paper Click to read the abstract
L.Buglione & A.Abran, Improving Measurement Plans from multiple dimensions: Exercising with Balancing Multiple Dimensions - BMP, 2005, reprinted in "Projects & Profits", July 2006, ICFAI University Press, pp.11-16
Buglione L., Gencel G. & Efe P., Suggestions for Improving Measurement Plans: a BMP application in Turkey, Proceedings of IWSM / Metrikon 2006, Potsdam (Germany), November 2-4, 2006, Shaker Verlag, ISBN 3-8322-5611-3, pp. 203-227 Click to read the abstract
Cuadrado-Gallego J.J., Buglione L. & Rodriguez D., Suggestions for Improving Measurement Plans: A BMP application in Spain, Proceedings of the 4th Software Measurement European Forum (SMEF 2007), Rome (Italy), May 9-11 2007 Click to read the abstract


[Bio Sketch] [Misurare il Software] [Publications] Home Page [QEST & LIME models] [Presentations] [Links]

Last update: March 14, 2007
Previous update: October 14, 2006
Created: May 1, 2005

Hosted by www.Geocities.ws

1