Using Multi-Attribute Decision Making

to Define, Prioritize and Manage Software Requirements



By Haoyu Wang

SKEMA Business School

Paris, France



In the development process of software projects, requirement management runs through the entire life cycle of software projects. In software project management, requirements engineering is the first step in software development, which is a key step and the most difficult step. The quality in requirement management directly affects the quality of software and even the success or failure of software projects. This determines that the project team must have a requirement management strategy and effective implementation.

This paper uses multi-attribute decision making to define and prioritize several behavioral development models that manage software requirements and will help users understand the most critical criteria they should choose (from the customer’s point of view), the advantages and disadvantages of various methods in different aspects. This should allow a team to achieve barrier-free communication, error-free delivery.

Keywords: Specification by Example (SBE),Requirements, information technology, Software, System product, agile, behavioral development models


As IT systems become more and more critical competitive factors in the industry, the scale of the IT project is more and more prominent, involves the organization level is higher and higher, the risk of the project is becoming more and bigger. As one of the fastest growing areas in the world today, the failure of the project also represents a large amount of capital loss and time cost and the waste of labor costs. According to the PMI industry statistics, ‘the average waste of IT industry project expenditures in 2017 was as high as $78 million per $1 billion. Among them, the average waste was the highest regarding geographical distribution and European project expenditure, and $131 million was wasted for every $1 billion.[1]

Then in the field, there are many examples of companies being dragged down by the IT project or the returns were much lower than expected. The data in Figure 1 is the result of the IT project in earlier years. We can see that only 39% of IT projects were successful in that year. This figure means that more than 50% of the IT projects did not achieve the expected results.

Figure 1: Software development project results[2]

Then the software that has been successfully developed must be no problem, without the doubt? However, it is not. According to statistics, there are still many customers not satisfied with the final delivery software. And after customers use the software for two months, the percentage of dissatisfaction can double. This figure is a horrifying number and lets us take a look at the situation of industry practitioners. In the project development work, many programmers have not understood the target results. With this requirement, people can start development. However, the reality is that people do not know what to do, but they continue to develop. When the project is postponed, or there is a problem with the project test, time and money have been lost, and Inconsistent with the original intention.

The technology is advanced, the team is strong, and the products have a market. However, it still failed because – “delivered the wrong software.” However, for a project that delivers a piece of software, “delivering the wrong software” is often the origin of the dispute. However, it is also a difficult problem to solve. For the owner, he is paying for a “right software.” However, there are often hardships for the developers who are entrusted, because they are given instructions to “act the software according to the precise functional description of the owner.” “Correct or not” is not their ultimate responsibility. Whether or not “correct” usually has to wait until after the online, the customer can know according to the user’s actual measurement response (although the developer is often the development stage can usually guess the group of people who failed). However, this is never within the responsibility of the contract.

STEP 1: Problem Definition

How did this happen, why did it become like this? Let us have a look at some of the features of IT project to help us sort out the problems.

In the IT project, the following points are different from the general project:


To read entire paper, click here


Editor’s note: Student papers are authored by graduate or undergraduate students based on coursework at accredited universities or training programs.  This paper was prepared as a deliverable for the course “International Contract Management” facilitated by Dr Paul D. Giammalvo of PT Mitratata Citragraha, Jakarta, Indonesia as an Adjunct Professor under contract to SKEMA Business School for the program Master of Science in Project and Programme Management and Business Development.  http://www.skema.edu/programmes/masters-of-science. For more information on this global program (Lille and Paris in France; Belo Horizonte in Brazil), contact Dr Paul Gardiner, Global Programme Director [email protected].

How to cite this paper: Wang, H. (2018). Using Multi-Attribute Decision Making to Define, Prioritize and Manage Software Requirements, PM World Journal, Vol. VII, Issue XII (December). Available online at https://pmworldjournal.net/wp-content/uploads/2018/12/pmwj77-Dec2018-Wang-define-prioritize-manage-software-requirements.pdf

About the Author

Haoyu Wang

Paris, France


Haoyu Wang
, Chinese, 23 years old, major in Project and Programme Management & Business Development (PPMBD) at Skema Business School in Paris, France. He has a background in business and IT education.

Before coming to France, Haoyu completed his studies at ZhaoQing College in China and hold a bachelor’s degree in the Internet of Things. Haoyu has been working for ‘Network Information and Security Research Institute’ (China) and ‘Internet of Things Collaborative Innovation Center Embedded Systems Application Institute’ (China) for many years. During this time, he assisted in the completion of the ‘SMS-based smartphone virus propagation analysis and simulation system research’ and the ‘Unmanned Helicopter Quadrotor,’ ‘WiFi student attendance system.’ From 2015 to 2016, Haoyu hosted the provincial assignment tasks ‘climbing plan’ and ‘Challenge Cup.’ Finally, the task is perfectly delivered with ‘small block type wireless LED display design’ and ‘smart home remote monitoring system development.’ After that, he assisted the professor in publishing an article entitled “Spi / Uart and Zigbee Protocol Conversion Module Design” in the IoT Journal. In 2016, he became an executive intern at Hong Kong AXA Insurance Company. Participate in the ‘fund selection plan and solve the difficulties and countermeasures of financial planning in mainland China.’

Haoyu has the potential to be a promising project manager. At the same time, he has great enthusiasm and interest in the IT field.Haoyu is now working on a master in sciences in Skema business school.He lives in Paris now and can be contacted at [email protected]  or https://www.linkedin.com/in/haoyu-wang-0a21b1161/


[1] Pulse of the Profession | Project Management Institute. (November 11). Retrieved from https://www.pmi.org/Pulse

[2] Sweeney, M. (2015). 3 Reasons Why Software Development Projects Fail – Clearcode Blog. Retrieved from https://clearcode.cc/blog/why-software-development-projects-fail/