A CMMI-compliant requirements management and development process

Requirements Engineering has been acknowledged an essential discipline for Software Quality. Poorly-defined processes for eliciting, analyzing, specifying and validating requirements can lead to unclear issues or misunderstandings on business needs and project’s scope. These typically result in customers’ non-satisfaction with either the products’ quality or the increase of the project’s budget and duration. Maturity models allow an organization to measure the quality of its processes and improve them according to an evolutionary path based on levels. The Capability Maturity Model Integration (CMMI) addresses the aforementioned Requirements Engineering issues. CMMI defines a set of best practices for process improvement that are divided into several process areas. Requirements Management and Requirements Development are the process areas concerned with Requirements Engineering maturity. Altran Portugal is a consulting company concerned with the quality of its software. In 2012, the Solution Center department has developed and applied successfully a set of processes aligned with CMMI-DEV v1.3, what granted them a Level 2 maturity certification. For 2015, they defined an organizational goal of addressing CMMI-DEV maturity level 3. This MSc dissertation is part of this organization effort. In particular, it is concerned with the required process areas that address the activities of Requirements Engineering. Our main goal is to contribute for the development of Altran’s internal engineering processes to conform to the guidelines of the Requirements Development process area. Throughout this dissertation, we started with an evaluation method based on CMMI and conducted a compliance assessment of Altran’s current processes. This allowed demonstrating their alignment with the CMMI Requirements Management process area and to highlight the improvements needed to conform to the Requirements Development process area. Based on the study of alternative solutions for the gaps found, we proposed a new Requirements Management and Development process that was later validated using three different approaches. The main contribution of this dissertation is the new process developed for Altran Portugal. However, given that studies on these topics are not abundant in the literature, we also expect to contribute with useful evidences to the existing body of knowledge with a survey on CMMI and requirements engineering trends. Most importantly, we hope that the implementation of the proposed processes’ improvements will minimize the risks of mishandled requirements, increasing Altran’s performance and taking them one step further to the desired maturity level.

Escolha de indicadores para um relatório

This dissertation is aimed at helping organizations that implemented a Business Intelligence (BI) system without documenting to identify the reasons for the indicators choice either in the conception phase of the project or other. The example taken to present the methodology is a fictitious case study of an organization named BestBread. The aim of this dissertation is to demonstrate not only the necessary indicators in a report but also to describe why they are needed through a business goal representation. This dissertation approach focus mainly in using two methodologies, a simplified notation of the Business Intelligence model (BIM) and a systematic approach that aims to justify BI indicators through modelling report goals. This approach provides guidance to organizations that already implemented a BI tool by presenting a method to compare intuitive and systematic selection of indicators with the BI system existing indicators. The approach is applicable to define in a report its significant indicators. The steps needed to be executed are the following: 1- Model business goals; 2- Select indicators through an intuitive perspective; 3- Verify the indicators existence identified in the intuitive perspective; 4- Select indicators through a systematic perspective; 5- Verify the indicators existence identified in the systematic perspective; 6- Make a global comparison. The dissertation approach allowed an easier way to identify and explain the purpose of indicators to be used in a report. Also, the methodology presented could help the BI deployment phase to be quicker since users would be able to visualise through the representations the evaluation that the indicators could evoke in their business goals. Therefore, it could improve the use of the BI tool, its acceptance and maybe even users’ satisfaction with the tool.