EVALUATION OF PMBOK AND SCRUM PRACTICES FOR SOFTWARE DEVELOPMENT IN THE VISION OF SPECIALISTS

In project management, the challenge for software development is to achieve success for the proposed projects, using methods such as PMBOK and Scrum. Knowledge of the advantages of these methods are critical success factors for product development. Therefore, the proposal of this study was to verify the perception of specialists of the area of software development on practices of project management. The used methods in this study were the bibliographic, exploratory and qualitative research, with the construction of a questionnaire with 14 items on the advantages of project management practices of various natures, size and complexity, which were applied in 90 specialists. The results of the research demonstrated that all the experts agreed with the advantages of the project management practices, identified based on the literature review, for software development, thus validating the proposed items of the questionnaire. It is recommended for future researches the accomplishment of case studies that explore practical models of evaluation of the use of the practices studied in the scope of software development. It is important in the future studies that metrics and indicators be drawn for each of the advantages cited in the present study


INTRODUCTION
The context in which organizations around the world operate is becoming increasingly complex and competitive. The speed at which cultural, political, economic, and social changes are taking place naturally presents opportunities and threats that impact the achievement of the organizational mission, whether of public administration organizations, business entities or non-profit entities (HORTA, 2009).
Organizations often experience partial or total failure with regard to the scope, quality, timing, and cost planned in their projects and programs. In spite of the existence of vast literature on the subject and intense work of international organizations such as the International Project Management Association (IPMA), the Project Management Institute (PMI) and the Association for Project Management (APM), which bring together professionals of different backgrounds and performance in several countries of the world (BARRETTO et al., 2008), it is still possible to verify a large percentage of project failure, which according to Shenhar and Dvir (2007), in the field of organizational strategies, these initiatives have failed suggesting that Projects Management such as discipline and methodology has been largely neglected. Thomsett (2001) states that a project can address a number of quality considerations, and these vary according to how each stakeholder sees the project.
The development team cares about the technical aspects while business analysts and project managers care about managerial aspects. Among the technical aspects can be mentioned the specification of functionalities, the data model, the test plan and the project documentation, etc. The managerial aspects are those dealing with aspects such as resource allocation, project priorities, risk analysis, return on investment and cost estimates, etc. In this way, the selected methodology must support the interests of both groups.
According to Cockburn (2000), every project needs its own methodology and each methodology has a specific weight. The weight, according to the author, is composed of two components: the size that corresponds to the number of elements in the methodology and the density that corresponds to the "rigour" with which these elements must be realized. http://www.ijmp.jor.br v. 8, n. 5, Special Edition IFLOG 2016ISSN: 2236 Practices involving identification and prior treatment regarding risks and uncertainties that may affect project performance are of fundamental importance in the business environment, where pressure for results is one of the major factors involved because of the competitive nature of this scenario. (DEY; CLEGG; BENNETT, 2010).

INDEPENDENT JOURNAL OF MANAGEMENT & PRODUCTION (IJM&P)
One challenge is to make every software development project achieve success for it. In this sense, the search for a proper methodology for each project becomes a fundamental issue for their success (COCKBURN, 2000).
Within this context, the correct choice of methodology in software development becomes a critical factor for success in this segment.
Thus, the proposal of this study was to verify the perception of specialists on project management practices, identified based on the literature review, for the area of software development.

LITERATURE REVIEW
This section presents concepts about the software development process and its relationship with project management.

Software Development Process
According to Sommerville (2004), a software development process can be defined as a set of activities, methods, practices and transformations that people employ to develop and maintain software and associated products (for example, project plans, project documents, design, code, test cases, user manual).
For Pressman (2006), the software process is the set of partially ordered activities that a project must follow to perform some task. This task should aim to achieve a certain goal and is usually, associated with the development of one or more products. Some models of software development can be cited:   • Fourth-generation techniques, this paradigm focuses on the ability to specify software close to the natural language. This model uses the following steps (1) Requirements collection, (2) Project strategy, (3) Implementation using 4th Generation Techniques and (4) Test (PRESSMAN, 2006).

INDEPENDENT JOURNAL OF MANAGEMENT & PRODUCTION (IJM&P)
Software development processes act as guides for the execution of all projects within an organization.

Project Management
To better understand project management it is fundamental to understand the concept of design. Schwalbe (2002) defines a project as a temporary enterprise with the objective of producing a single product or service.
The definition given by Dinsmore and Silveira Neto (2005, p

.1)
A project is a temporary effort made to create a unique product or service, unlike any other, of all other products and services, with beginning and end, defined, which uses resources, is managed by people and obeys the parameters of cost, time and quality. http://www.ijmp.jor.br v. 8, n. 5, Special Edition IFLOG 2016 ISSN: 2236-269X DOI: 10.14807/ijmp.v8i5.598 In this way, a project can be defined as a non-repeatable enterprise, with a clear and logical sequence, with defined beginning and end and with the specific and defined objective. In turn, being conducted by people respecting the predefined parameters of cost, time, resources involved and quality, as stated by Martins In addition, the PMBOK has five groups of project management processes that are interdependent. Thus, in process groups, the outputs of a process become inputs to another process or are deliveries of the project. The PMI also describes that these activities are overlapping and occur at varying levels of intensity throughout the project. Thus, for projects that are divided into phases, the process groups interact within a project phase and may cross several phases of the project, as shown in Figure 1.   DOI: 10.14807/ijmp.v8i5.598 For scenarios characterized by constant changes of requirements and the need to perform deliveries in a short time, it is recommended to use agile methods for product development.
In general, agile methods divide development into several short iteration cycles, and at the end of each cycle deliveries are made, so that "[...] the customer (internal or external) receives a version that adds value to their business" (DANTAS, 2003, p.37). In this way, developers can track requirements changes at the beginning of each cycle. This provides customer feedback to the development team, which reduces project risk.
The principles of lean manufacturing, implemented by Honda and Toyota companies, and Takeuchi and Nonaka (2004) and Senge (1990) knowledge management strategies strongly influenced these methods.   http://www.ijmp.jor.br v. 8, n. 5, Special Edition IFLOG 2016ISSN: 2236 The legend (a) presents the waterfall model having its activities distinguished in grayscale, showing that it is a cycle with long duration. In (b), the cycles are shorter. The evolution of iterative development to agile methods is observed in Scrum in (c), where analysis, design, coding and testing activities are combined in short periods throughout the project.

METHODOLOGY
In this study, the bibliographic research in the SciELO and Scopus bases was  Table 1 shows the 13 criteria that were used in the questionnaire of this research, in order to obtain the users' perceptions of the process regarding their degree of agreement. The criteria were consolidated using similarity of the nomenclature used by different authors.
Greater involvement and collaboration among team members. Visibility of activities by all team members resulting from the use of Scrum practices.

Visibility of project activities by all team members.
High agility for requirement adjustment at design time.
Ease of adjustment of requirements at design time due to short deliveries of product running. Feedback speed Scrum practices can contribute to project closure faster.
Closure of the project more quickly, mainly due to feedback generated by the short deliveries. Thus avoiding rework. Scrum practices can contribute to lower project costs.
Decrease in project costs mainly due to the rapid feedback generated by short deliveries.
A Likert scale was used to collect data on the degree of agreement of the research participants regarding the proposed criteria, according to Table 2. This scale was adopted based on Likert (1932) and Costa (2010). The questionnaire was applied to 90 specialists related to the development of new software products and that are located in companies of Rio de Janeiro and São Paulo, belonging to the private initiative or to the government.
The data obtained were treated using descriptive statistics techniques, whose objective, according to Larson and Farber (2007) is "to make it easier to understand the data and to see trends, averages and variations in them".

ANALYSIS OF RESULTS
This section deals with the analysis and results of this study and is organized in the following subsections: identification of respondents and analysis of the advantages of using project management practices using PMBOK and Scrum methods in the area of software development.

Identification of Respondents
The characterization of the respondents was obtained based on the following items: 1. Segment of the institution where you work 2. Practice Area
Graphic 1: Segment of the institution where you work As for the first item, it was verified that the highest incidence of responses is concentrated in the area of Technology and Innovation with 42%, the second most selected option with 27% was industry, construction and engineering, the third most selected option was Education and scientific Development with 19%. The trade and services option obtained only 2%, while the option "others" reached 10%.

Graphic 2: Practice Area
As for the second item, the most answered that works on the Information Technology or Telecommunications sector, totaling 68%. The second option with expressive volume of responses was Strategy, Process, Planning, Management or http://www.ijmp.jor.br v. 8, n. 5, Special Edition IFLOG 2016ISSN: 2236 Project Management Office (PMO) with a total of 21%. The other options and engineering obtained 2% each, while technological development obtained 7%.

INDEPENDENT JOURNAL OF MANAGEMENT & PRODUCTION (IJM&P)
Graphic 3: Experience time in the area.
As for the third item, it can be verified that the absolute majority of the respondents has more than 10 years of experience, being a point that can be considered very positive, since it is a survey of opinion and perception. Altogether 62% of the respondents have more than 10 years of experience followed by 31% of respondents in the range of 5 to 10 years. The remaining 7% have less than 5 years' experience.

Analysis of the advantages of using PMBOK and Scrum practices
In the Table 3, the results obtained, as well as the evaluated items of the questionnaire applied to specialists in the area of software development, are presented. According to the data presented, it was verified that the majority of the respondents totally agreed with eight of the advantages and agreed partially with six advantages. This confirms the total convergence of the specialists in relation to the data from the literature review.
When analyzing the advantages of using project management practices in software development, it was found that the majority of the respondents, 80%, agreed that the PMBOK offers a detail of planning and control, and about 88% agreed that both the PMBOK and Scrum practices promote improved communication among team members.
It is worth mentioning that 89% of the respondents considered the occurrence of learning in a short period provided through retrospectives of the Scrum iteration.
On the ease of use of Scrum a total of 83% agreement was obtained, while 66% of the respondents pointed the greater documentation provided by the PMBOK.
In addition, 82% of respondents agreed on the advantage of having product running in short periods in Scrum.
It was also evidenced that 74% of the specialists answered that they agree with the increase of motivation of the team resulting from the use of Scrum practices.
In relation to the productivity increase of the Scrum practices a total of 84% was obtained, while 81% of the respondents agreed with the increased control of the project activities resulting from the daily meetings resulting from the Scrum practices; The results also indicated that 83% of respondents agreed on improved collaboration between those involved in using Scrum practices. In addition, it was noted that: 87% of respondents agreed that Scrum practices provide visibility of activities by all team members; 75% of respondents agreed that Scrum provides high

INDEPENDENT JOURNAL OF MANAGEMENT & PRODUCTION (IJM&P)
http://www. ijmp.jor.br v. 8, n. 5, Special Edition IFLOG 2016ISSN: 2236 agility for requirement adjustment at design time; 76% of respondents agree that Scrum practices can contribute to project closure more quickly. In relation to the reduction of project costs resulting from the use of Scrum practices, a total of 64% of agreement was obtained.
It was observed that the degree of agreement in all items was expressive, that is, above 60%, which indicates that these items have a positive evaluation of the experts regarding the proposed study.

CONCLUSION
Through the literature review, it was possible to identify 14 advantages related to the project management practices in several areas of application, which were submitted to the evaluation of specialists in the area of software development, to verify the degree of agreement of these experts on these criteria.
Based on these results, the present study is considered as an important contribution tool for the development of scientific knowledge about the advantages of the project management practices in software development.
Therefore, the main contribution that can be drawn from this work is a list of advantages identified based on the literature review on PMBOK and Scrum project management practices in software development. This study allowed the validation of these criteria identified by the specialists of the software development sector, corroborating the importance of this study.
Finally, the recommendation for future research is to carry out empirical studies of the case study that explore practical models of evaluation of the use of the practices studied in the scope of software development. It is important in the future study that metrics and indicators should be drawn for each advantage cited in the present study.