Abstract
The advancement of knowledge in the field of project management has consisted of the development of an increasing number of processes, tools, and techniques. Decisions on which of these resources will be used in order to accomplish project objectives in the most efficient way are contingent upon specific project features such as its objectives, resources, and context. One of the most challenging of such features is the project's complexity. For most authors, the complexity of a project can be measured in terms of its degree of novelty, its interdependencies, and the technologies involved, among other variables. For some authors, the idea of complexity entails a subjective connotation.
This article is a compilation of the main results and conclusions of a master's degree dissertation. The objective of this study is to contribute to the knowledge on the concept of project complexity, on the dimensions that complexity consists of, and to identify possible relations between project complexity and the frequency of use of specific project management processes and techniques. In addition, this study also aimed at identifying dimensions related to the complexity of projects, from the perspectives of their managers, in accordance with their experience. A questionnaire was submitted to 313 project managers. The data analysis indicated that the complexity of projects managed by the participants can influence the frequency of use of processes and techniques related to project planning and people management.
Keywords: management process; project complexity; management techniques; complexity dimension.
Introduction
The term “project” is commonly used with different meanings. Boutinet (1993) refers to projects as figures ubiquitous in social life, defined by three core characteristics: uniqueness, materialization of the objective, and identity. Project Management Institute (PMI) (2008) adopts the concept that projects deliver products, services, or results that are exclusive, non-repetitive, or unique. Projects are also defined as temporary endeavors that start with an idea and end with the delivery of the materialization of this idea.
In the same way, there is no unique, universally accepted definition about what project management is (Smyth & Morris, 2007). It can be explored in several perspectives, depending on different factors, such as culture and values, interests, and strategic goals. According to Smyth (2009), this characteristic demands the use of several management tools.
Project management involves decisions about planning, organization, execution, control, and conclusion (PMI, 2008). The decisions apply to the variables (or areas of knowledge) present in any project, such as: integration, scope, time, costs, communication, people, risks, and so on. To each one of these expertise areas there are corresponding quantitative and qualitative procedures that assist in the process of decision-making about how to manage the project. These procedures are called processes, tools, or techniques.
The most complex projects of the 1960s, such as the program designed to land humans on the Moon and bring them safely back to Earth (Apollo), were important not only for the products or results that they generated, but also because of the tools and techniques that were developed for their management (Shenhar & Dvir, 2007; Shtub et al., 1994). According to Shtub et al. (1994), carrying out projects with skills not available in advance is one of the factors that determined the development of new methods for planning, organization, and controlling, which would be the essence of project management.
For Maximiano (2002), although there may be significant differences between different projects, the management principles used in their management are the same. Project management is a technique or set of techniques that can be applied to specific situations, according to the intrinsic nature of the situation and conscious choice; thus, knowledge and practices of project management do not apply uniformly to all projects (PMI, 2008).
In general, projects are implemented through management processes. The fourth edition of A Guide to the Project Management Body of Knowledge (PMBOK® Guide) (PMI, 2008) contains 42 project management processes distributed among five Process Groups: Initiating, Planning, Executing, Monitoring & Controlling, and Closing. Le Bissonnais (2003) says that the notion of process is independent of the size and complexity of the project and serves to simplify the management by identifying the elements, tools, and resources needed for the management.
The study conducted by Besner and Hobbs (2010) aimed to identify empirically the set of tools used in the practice of project management. By conceptualizing the multidisciplinary field of the development of projects, the researchers could build a practical overview of project management. They identified the existence of classes of standards of practices of project management. Research has shown that professionals in project management use tools and techniques in groups, which can be identified and studied empirically, as they are known and used in daily activities. Among their conclusions, the researchers (Ibid.) indicate that sets of project management tools are used in many different contexts, each one with its particular management problems, for which the practices of management projects have been adapted and the management skills developed for their use. Additionally, they identified variations among the patterns of practices due to differences in the types of projects. The list of project management processes and techniques adopted by this research is presented in Table 1.
Table 1: Project management processes and techniques.
| Processes | Process group | Reference |
| Project Charter | Integration | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mulcahy (2005); Streun (2006) |
| Kick-off meeting | Communications | Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mulcahy (2005); Le Bissonnais (2003) |
| Change control | Integration | PMI (2008); Mulcahy (2005); Streun (2006); Cleland (1999); Le Bissonnais (2003) |
| Identification of project requirements | Scope | PMI (2008); Besner & Hobbs (2008); Mulcahy (2005); Peels (2006); Le Bissonnais (2003) |
| Project scope statement | Scope | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mulcahy (2005); Mepyans-Robinson (2006) |
| WBS - Work breakdown structure | Scope | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mulcahy (2005); Turner & Cochrane (1993); Peels (2006); Mepyans-Robinson (2006); Le Bissonnais (2003) |
| WBS Dictionary | Scope | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mulcahy (2005); Mepyans-Robinson (2006a); Cleland (1999) |
| Resources estimate | Time | PMI (2008); Houston (2006) |
| Project Schedule | Time | PMI (2008); Mulcahy (2005); Houston (2006); Cleland (1999) |
| Cost estimate | Costs | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mulcahy (2005); Peels (2006); Abdomerovic (2006); Cleland (1999) |
| Project budget | Costs | PMI (2008); Mulcahy (2005); Abdomerovic (2006) |
| Earned value analysis | Costs | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Ellis Jr (2006); Lambert (2006) |
| Quality management plan | Quality | PMI (2008); Peels (2006); Le Bissonnais (2003) |
| Project organization chart | People | PMI (2008); Maximiano et al. (2011a, 2011b); Peels (2006); Le Bissonnais (2003) |
| Responsibility assignment matrix | People | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Towe (2006); Cleland (1999) |
| Team-building | People | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mulcahy (2005); Dinsmore (2006); Le Bissonnais (2003) |
| Project communication plan | Communication | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mepyans-Robinson (2006b) |
| Stakeholders management | Communication | PMI (2008); Mepyans-Robinson (2006b); Englund (2006) |
| Project performance reports | Communication | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mulcahy (2005); Peels (2006); Mepyans-Robinson (2006b); Lambert (2006); Le Bissonnais (2003) |
| Maintainance of project registers | Communi-cation | PMI (2008); Besner & Hobbs (2008); Mulcahy (2005); Peels (2006) |
| Matrix of risk analysis | Risks | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Hillson (2006) |
| Strategy to risk responses | Risks | PMI (2008); Maximiano et al. (2011a, 2011b); Mulcahy (2005); Hillson (2006) |
| Procurement management plan | Procurement | PMI (2008); Le Bissonnais (2003); Mulcahy (2005); Woolf (2007) |
| Contracts management | Procurement | PMI (2008); Maximiano et al. (2011a, 2011b); Mulcahy (2005); Edwards (2006) |
| Register of lessons learned | Knowledge | Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mulcahy (2005); Edwards (2006) |
| Statement of acceptance issued by the client. | Integration | PMI (2008); Maximiano et al. (2011a, 2011b); Besner & Hobbs (2008); Mulcahy (2005) |
Archibald (2003) states that the objective of a project, or the type of product that it generates, is one of the factors that defines how a project will be managed, differentiating it from others. However, even if different projects have similar goals, they will differ from each other, depending on the context in which they are executed and on the degree of complexity they represent to their managers. Thus, according to Shenhar and Dvir (2007), the context can be evaluated according to the complexity, uncertainty, and the degree of familiarity with the organization's projects.
Among the characteristics of projects, complexity is one that has received increasing attention. Williams (1999) affirms that, in general, there are two basic reasons for this phenomenon: extensions or improvements of previous generations of products (e.g., additional features and greater inter-relationships) and increasing project restrictions with respect to deadlines and expectations of rapid delivery of products. On the other hand, the results of the research conducted by Xia and Lee (2004) indicate that the complexity is associated with delays, cost overruns, restrictions of system functionalities, and reduction of user satisfaction.
The concept of systems is used to define important components of projects complexity related to interconnections and interdependences of organizations and technologies (Baccarini, 1996; Geraldi, 2007, Geraldi & Adlbrecht, 2007; Remington et al., 2009). According to Shtub et al. (1994), complexity is at the root of the concept of projects, because the factors that determine the realization of a project include the complexity never seen before in its design, development, and implementation.
Remington et al. (2009) agree that complexity is important for the management of projects because of difficulties associated with decision-making and achievement of goals. However, they indicate a lack of operational definitions for complex projects. Geraldi (2007), Adlbrecht and Geraldi (2007), and Williams (1999) believe that it is necessary to look at the project and its problems holistically, considering not only the specific components, but also its effects. In general, the concept of complexity of projects has been discussed from different perspectives and there are different definitions with regard to the concepts of difficulty, uncertainty, and complexity. It is important to consider, however, that despite the established concept of complexity in other areas, such as physics and biology, the definitions adopted by this study are restricted, in theory, to the knowledge area of management and, more specifically, the management of projects.
An important issue on studying the complexity of the projects is related to the dimensions that generate or influence it. Baccarini (1996), Williams (1999), and Fitsilis (2009) proposed a classification based on two types of complexity: the organizational and technological, which are operationalized in terms of differentiation and interdependence. Tatikonda and Rosenthal (2000) refer to the concept of project complexity as relating to the novelty of the product, of its development process and performance objectives, and of its technological interdependence and difficulty.
Geraldi (2007) and Geraldi and Adlbrecht (2007) proposed an approach that considers that complexity management should be the result not only of unique characteristics of complexity, but of a pattern. Based on this idea, they developed the concept of “pattern complexity”; i.e., minimal manageable “space” of complexity. They defined three main types of complexity: Complexity of Faith (related to uncertainty), Complexity of Fact (referring to the amount of interdependent and concurrent information), and Complexity of Interaction (with respect to the interfaces between systems, people, and places).
Remington et al. (2009), based on the analysis and classification of the factors related to the projects’ complexity, defined two groups of categories: dimension of complexity (characterize the nature or origin of complexity) and factor of severity (to what extent is a problem). In an even broader way, Fitsilis (2009), Baccarini (1996), Geraldi and Adlbrecht (2007), and Remington et al. (2009) believe in subjective connotation, which would be the main reason for the difficulty in understanding and dealing with the project's complexity.
Although there are many studies on the issue of complexity in projects, it was not possible to identify a model that could indicate which dimensions are responsible, in particular, for the complexity of certain types of projects. For the purpose of this study, the term “dimension” was adopted as a substitute for the terms “factor,” “component,” or “attribute” when relating to items that contribute to project complexity. Table 2 shows the main dimensions of complexity found in the literature.
Table 2: Dimensions related to the complexity of projects.
| Dimension of complexity | Reference |
| - Uncertainty about the scope of the project | Tatikonda & Rosenthal (2000); Maximiano (2002; 2010); Turner & Cochrane (1993); Remington et al. (2009); Scranton (2008); Williams (1999); Baccarini (1996); Shtub et al. (1994). |
| - Uncertainty about the product of the project | Tatikonda & Rosenthal (2000); Maximiano (2002; 2010); Turner & Cochrane (1993); Remington et al. |
| (2009); Scranton (2008); Williams (1999); Baccarini (1996); Shtub et al. (1994). | |
| - Significant change in the scope of the project during its implementation | Geraldi (2007); Geraldi & Adlbrecht (2007); Remington et al. (2009); Williams (1999); Fitsilis (2009); Turner & Cochane (1993). |
| - High difficulty to achieve performance goals | Williams (1999); Maximiano (2002); Tatikonda & Rosenthal (2000); Baccarini (1996); Fitsilis (2009). |
| - High number of stakeholders with influence on the project | Williams (1999); Remington et al. (2009); Fitsilis (2009). |
| - High interdependence between firms involved in the project. | Baccarini (1996); Williams (1999); Geraldi (2007); Remington et al. (2009); Maximiano (2010); Fitsilis (2009). |
| - Novelty of the technology | Baccarini (1996); Tatikonda & Rosenthal (2000); Fitsilis (2009); Williams (1999); Geraldi (2007); Remington et al (2009). |
| - High interdependence between the technologies | Geraldi (2007); Geraldi & Adlbrecht (2007); Remington et al. (2009); Williams (1999); Remington (1996); Pollack, 2007; Baccarini (1996); Fitsilis (2009); Tatikonda & Rosenthal (2000); Ireland (2007); Maximiano (2010). |
| - High multidisciplinary | Baccarini (1996); Maximiano (2002); Geraldi (2007); Geraldi & Adlbrecht (2007); Fitsilis (2009). |
| - Large number of different activities to be performed | Baccarini (1996); Williams (1999); Fitsilis (2009). |
The Method
This research aimed to contribute to the knowledge of project complexity, the dimensions that complexity consists of, and to identify the project management processes and techniques that are most frequently used in complex projects. It can be classified as non-experimental (Sampieri et al., 2006) and predominantly exploratory (Selltiz et al., 1974). The research had the participation of professionals working in Brazil as project managers, not limited to specific types of products or industry sectors. The sampling was non-random, by convenience (Fávero et al., 2009). The contact with the project managers was made by accessing discussion groups on the issue of project management that are established in social networks, such as Yahoo Groups®, Google Groups™ and LinkedIn®. The survey instrument was an electronic questionnaire (Vasconcellos & Guedes, 2007; Kalantari et al., 2011) applied through a website (QuestionPro™).
Statistical techniques were used for the data analysis, such as: descriptive statistics (Martins & Theóphilo, 2007; Sampieri et al., 2006; Fonseca et al., 1985), Spearman's correlation coefficient (Sampieri et al., 2006; Martins & Theóphilo, 2007), factorial analysis (Hair et al., 2005; Sampieri et al, 2006; Kim & Mueller, 1983), the Kruskal-Wallis test (Martins & Theóphilo, 2007; Fávero et al., 2009), and the multiple comparison test (Pontes & Corrente, 2001). The application of an electronic questionnaire obtained the participation of 313 professionals. The profiles of the participants are characterized by the experience in project management, considering that 57% of the 313 respondents have more than 6 years of experience in project management; in addition, 33% have professional certifications in project management.
The Results
The projects managed by the respondents in the last 12 months were distributed, in percentage terms, by the degree of complexity perceived by them. Table 3 shows the distribution of the projects managed among the different levels of complexity, on average percentage.
Table 3: Descriptive statistics: Distribution of projects by level of complexity.
| Without complexity | Low complexity | Medium complexity | High complexity | Very high complexity | ||
| N | Valid | 303 | 303 | 303 | 303 | 303 |
| Mean | 5.3% | 16.0% | 29.4% | 30.2% | 19.1% | |
| Median | 0% | 10% | 30% | 25% | 10% | |
| Mode | 0% | 0% | 50% | 20% | 0% |
The data presented in Table 3 show that, on average, 19% of the projects managed were classified as of “very high complexity.” If this percentage is added to the percentage of projects rated as of “high complexity,” it will represent almost half of the projects managed (49%).
The survey identified which dimensions were considered by the respondents to be the most important contribution to the complexity of the projects managed during the last 12 months of work. For this purpose, it applied a question considering 10 different dimensions (see Table 2), which were evaluated by the respondents according to a Likert-type scale of 5 points, ranging from “without complexity” to “very high complexity.” The results of the analysis of the responses are presented descriptively in Figure 1 and Table 4.
Figure 1: Complexity dimensions and their intensity in the projects managed.
Table 4: Descriptive statistics: complexity dimensions-intensity in the projects managed.
| In general (projects within the last 12 months) | Definition of the products scope | Definitions of the project management processes | Projects required technology | Multidisciplinarity necessary to the projects | Number of activities required by the projects | Stakeholders with influence on decisions | Criticality of project goals | Risk level of projects | Changes in scope during the project execution | Interdependence among technologies | Interdependence among firms | ||
| N Responses | 298 | 296 | 293 | 298 | 297 | 300 | 300 | 300 | 298 | 300 | 299 | 293 | |
| Not applicable | 5 | 7 | 10 | 5 | 6 | 3 | 3 | 3 | 5 | 3 | 4 | 10 | |
| Mode | 3 | 3 | 3 | 3 | 3 | 3 | 3 | 4 | 3 | 4 | 3 | 3 e 4 | |
| Percentiles | 25 | 3 | 2 | 2 | 2 | 3 | 3 | 3 | 3 | 3 | 3 | 3 | 3 |
| 50 | 3 | 3 | 3 | 3 | 3 | 3,5 | 4 | 3 | 4 | 3 | 3 | 3 | |
| 75 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | 4 | |
The criticality of the goals and changes in the project scope (mode = 4) were considered by the respondents to be the highest intensities of complexity in the projects managed. More than 50% of projects were considered to be high or very high complexity regarding these dimensions. This result was not expected considering that project performance goals regarding time, cost, scope, and quality are cited only by a few studies in the literature; one of them is by Remington et al. (2009) and includes the high-level goals among the key themes of the complexity of the projects. The criticality of the project goals seems to exert pressure on project managers and causes problems to the management's performance, especially if they are too restrictive.
Changes in the project scope during the execution phase are indicated as an important contributor to project complexity by researchers such as Geraldi (2007); Adlbrecht and Geraldi (2007); Remington et al. (2009); Williams (1999); Fitsilis (2009); and Turner and Cochrane (1993). Frequently, these changes result in redefinitions, preplanning, and insecurity (PMI, 2008). The criticality of the goals and changes in project scope has a relationship of mutual impact, because the changes in scope may result in the goals being pressured. On the other hand, when the scope is changed, there will be re-planning and rework, which normally lead to increased costs and time, thus resulting in further pressure on the goals.
The interdependence between firms and the influence of stakeholders appear on one level slightly below in terms of complexity and seem to indicate that, in practice, the complexity is much more related to dimensions that are out of the action of the project manager, in the sense that it depends on others.
Regarding the project complexity dimensions on the opposite side of the list, the technology needed to develop the project is one of those evaluated with a relatively minor contribution. This finding was not expected, considering the emphasis placed on technology as an important contributor to the project's complexity, as highlighted by Baccarini (1996), Tatikonda and Rosenthal (2000), Fitsilis (2009), Williams (1999), Geraldi (2007), and Remington et al. (2009). The type of project managed and considered in the sample could be the reason for the relatively little importance given to this complexity dimension, because only a small percentage (about 20%) relates to projects of high technology.
The correlation analysis (Spearman's correlation coefficient) between the degree of complexity of the dimensions in the projects managed and the distribution of projects, in percentage terms, among the different degrees of projects complexity, are illustrated in Table 5.
Table 5: Relations between the dimensions of complexity and distribution of projects by
| Distribution of the projects managed by degree of complexity | ||||||
| Complexity dimensions | Without complexity | Low complexity | Medium complexity | High complexity | Very high complexity | High + Very high complexity |
| In general (for projects managed in the last 12 months) | -0.33 | -0.49 | -0.50 | 0.33 | 0.56 | 0.75 |
| Definition of the scope of the products of the projects | -0.24 | -0.35 | -0.31 | 0.30 | 0.41 | 0.53 |
| Definition of project management processes | -0.09 | -0.14 | -0.12 | 0.08 | 0.24 | 0.24 |
| Technologies required to develop the project | 0.00 | -0.05 | -0.08 | 0.09 | 0.10 | 0.14 |
| Diversity of expertise of the professionals (multidiscip-linarity) | -0.10 | -0.25 | -0.25 | 0.14 | 0.29 | 0.36 |
| Number of activities necessary for the project | -0.16 | -0.26 | -0.20 | 0.10 | 0.34 | 0.35 |
| Stakeholders | -0.15 | -0.17 | -0.15 | 0.08 | 0.25 | 0.30 |
| Criticality of the goals. | -0.11 | -0.22 | -0.22 | 0.13 | 0.26 | 0.34 |
| Level of risks | -0.13 | -0.19 | -0.23 | 0.11 | 0.36 | 0.36 |
| Changes in scope during the execution of the project | -0.09 | -0.19 | -0.17 | 0.11 | 0.30 | 0.29 |
| Interdependence / integration of technologies | -0.03 | -0.16 | -0.15 | 0.06 | 0.27 | 0.26 |
| Interdependence among the firms | -0.07 | -0.28 | -0.24 | 0.13 | 0.33 | 0.39 |
Considering only a small number and well-defined complexity dimensions, an expectation of this study was that it would be possible to characterize the complexity of the projects and its influencers. However, the data analysis only found a correlation between the percentage of projects managed with high or very high complexity and the project's complexity dimensions in general (r = 0.75), which indicates that the complexity of the projects considered in this research is probably not a function of a specific complexity dimension, but rather a set of dimensions. This conclusion is in line with other authors, such as Geraldi (2007), Geraldi and Adlbrecht (2007), Williams (1999), and Remington et al. (2009) who believe that one must look at the complexity of the projects holistically.
A second dimension that presented significant and positive correlation with respect to the complexity of the projects was the definition of the scope of the products of the projects (r = 0.53). Based on this, one can infer that the definition of scope is one of the biggest challenges faced by project managers when dealing with complex projects. One of the reasons may lie in the concentration of many features of complexity in the planning phase of the project scope because it covers the definitions of project objectives, identification of risks, and the assumptions, constraints, and goals to be achieved (PMI, 2008). It is also important to consider that, at this phase, the decisions and definitions about the necessary technologies, the interdependencies, and the goals occur.
Because of the number of the project management processes and techniques selected for this research (26 in total) a factorial analysis was performed in order to reduce it to a restricted number of variables that, even so, would enable the study without losing its capacity to explain the facts. The best solution obtained by the statistical treatment is composed of eight factors (Table 6).
Table 6: Factorial analysis of project management processes and techniques.
| Factor 1 | |
| - Project budget | |
| - Cost estimate | Cost Management |
| - Resource estimate | |
| Factor 2 | |
| - Matrix of risk analysis | Risk Management |
| - Strategy to risk responses | |
| Factor 3 | |
| - Contracts management | |
| - Procurement management plan | Procurement Management |
| - Quality management plan | |
| Factor 4 | |
| - Project schedule | |
| - Kick-off meeting | Planning |
| - Project scope statement | |
| Factor 5 | |
| - Responsibility assignment matrix | |
| - Project organization chart | People Management |
| - Project communication plan | |
| Factor 6 | |
| - Identification of project requirements | |
| - Team-building | Requirements Management |
| - Stakeholders management | |
| Factor 7 | |
| - WBS Dictionary | Scope Management |
| - WBS - Work breakdown structure | |
| - Earned value analysis | |
| Factor 8 | |
| - Maintainance of project registers | Controlling |
| - Project performance reports |
The Kruskal-Wallis test was performed in order to study the relationship between the project's complexity and the frequency of use of the project management processes and techniques. The results are presented hereafter (Table 7).
H0: C1 = C2 = C3 (C: level of projects complexity – without/low complexity; medium complexity; high/very high complexity)
Ha: at least two levels of complexity differ from each other regarding the frequency of use of the project management processes and techniques.
Table 7: Level of complexity and project management processes and techniques Kruskal-Wallis test (α =0.05).
| Costs | Risks | Procurement | Planning | People | Requirements | Scope | Controlling | |
| χ2 | 4.071 | 2.497 | 4.285 | 6.508 | 11.45 | 0.67 | 0.318 | 5.25 |
| p | 0.131 | 0.287 | 0.117 | 0.039 | 0.003 | 0.715 | 0.853 | 0.072 |
| Independent variable: in general, complexicity of the projects managed during the last 12 months. | ||||||||
The results of the Kruskal-Wallis test indicate that the frequency of use of the project management processes and techniques related to the factors of planning and people management varies in accordance with the complexity levels of the projects managed. The other processes and techniques did not present any significant difference in their frequency of use given the different levels of project complexity.
The data from the two relationships encountered were submitted to a multiple comparison test and the results are shown in Table 8.
Table 8: Multiple comparison tests for the different levels of complexity.
| Factor | In general (projects managed during the last 12 months) | n | Average scores |
| Planning | Without/low complexity | 16 | 91.31 |
| Medium complexity | 114 | 142.09 | |
| High/very high complexity | 135 | 130.27 | |
| Total | 265 | ||
| People Management | Without/low complexity | 16 | 81.81 |
| Medium complexity | 114 | 125.91 | |
| High/very high complexity | 135 | 145.05 | |
| Total | 265 |
Based on the results of the multiple comparison test (Table 8), it was possible to confirm that the people management and planning factors presented differences among the different levels of complexity on the projects managed. In other words, the projects classified as without/low complexity are less frequently used in relation to those classified as high/very high complexity.
The factor of planning is composed of three components: project schedule, kick-off meeting, and project scope statement. These processes and techniques of project management are among the most used by the research participants and are mainly associated with the early stages of a project, which is when some of the most critical decisions on projects occur. This phase is defined by Tatikonda and Rosenthal (2000) as complex, iterative, and often unstructured. The initial phase of a project is also characterized by the Complexity of Faith, as defined by Geraldi (2007) and Geraldi and Adlbrech (2007), when this is more intense, considering that the objectives and means are not well-defined, the team is new, and the problems are still unique.
Another possible reason for this relationship has to do with the criticality of the project goals. This dimension of the project complexity, as mentioned before, is considered the most important contributor to the complexity of the projects managed. In this specific case, project goals that are hard to achieve invariably involve time constraints. Thus, this may be another justification for having schedules among the most used components for the management processes used in the cases of complex projects. Based on these data, it is possible to know the level of complexity of a new project since its initial stage, when the product or objective of the project and how it will be achieved are revealed.
In spite of the project scope statement being one of the components in the factor of planning, the absence of a relationship between the frequency of use of components of other scope management processes and projects complexity was unexpected. This relationship would be justifiable considering that issues regarding project scope were indicated as important contributors to the project's complexity. But, on the other hand, the factor involving the WBS (Work Breakdown Structure) did not indicate a significant relationship. One possible explanation for this is the fact that, when analyzing the frequency of use of the different projects management processes and techniques, it was found that project schedule (83% of the sample) is used much more than WBS (47%). It may be because, sometimes, the project schedule is developed directly, as a substitute to the project scope statement, without having a WBS as one of its inputs.
Another factor that correlates to the level of complexity of the projects managed is people management. This factor is composed of the responsibility assignment matrix, the project organization chart, and the project communication plan. The people management, or human resources management, is not a subject widely discussed in the literature on the complexity of projects, but only on specific aspects such as interdependence and multidisciplinarity. However, Geraldi (2007) and Geraldi and Adlbrech (2007) addressed the soft skills of managers when talking about the complexity of interaction. For them, the number of physical interfaces, the schedule, time pressure, and stress were responsible for a greater complexity of interaction in the project, especially in its final phase. Baccarini (1996), in turn, points out that the differentiation and the interdependencies of the projects are managed by integration (i.e., communication, coordination, and control). Thus, the author feels it is acceptable to say that integration is the rational and essential role of project management.
Discussion
The first result that can be highlighted is the indication that, in general and in practice, the projects managed by the respondents have the performance goals as the most important dimension in terms of contribution to the project's complexity. This result is not emphasized in the literature, although it may be treated as a matter of difficulty instead of complexity. In spite of that, this issue deserves further studies on how the project goals influence the complexity, what the possible results are for the complexity generated in this situation, and what a project manager can do to deal with it.
In general, considering the restricted number of variables selected, there was hope that the analysis of the relationship between the dimensions of complexity of projects and the distribution of projects managed by their level of complexity would indicate a more precise characterization of the complexity in the specific case of projects. However, the correlation analysis did find significance in only two relationships. The first relationship was found between the percentage of projects managed with high or very high complexity and the complexity of projects related to the definition of the scope of the products of the projects. In this case, it seems reasonable to assume that this relationship is indicative that during the definition of project scope it would be possible to perceive its level of complexity, because, in its process, are considered or analyzed most of the characteristics of complexity of a project. With comparative higher significance, the other positive relationship caught our attention (i.e., the relationship between the complexity dimensions in general and the distribution of the projects managed according to their level of complexity). This result is indicative that it is plausible to consider that the project's complexity actually is not related to specific dimensions, but to a set of dimensions, reinforcing the idea of the holistic approach, which is mentioned by other researchers. Furthermore, in fact, it also seems to be plausible to consider personal aspects as part of the perception about the complexity of a given project. The project objectives can be considered very complex by one project manager and perceived differently by other project manager, even with the same background and resources, which is due to subjective connotation of the complexity. Considering this, maybe the real challenge is not the characterization of the complexity of a project, but the possible and best ways of dealing with it.
Finally, the results of the analysis on the effects of the level of complexity of the projects on the frequency of use of the processes and techniques of project management showed that there are relationships regarding the planning and the people management factors. According to these data, attention has to be given to the initial phases of the projects, which involve the scope and schedule decisions, and to the relationships, people management, and communications. On the other hand, no other significant relationships were found regarding the other factors of project management processes and techniques, even for the scope management factor. This result suggests that the projects, both complex and simple ones, are managed using the same processes and techniques of project management. But it is possible to consider that specific processes and techniques, other than those normally used to manage the project, are used in the case of the complex projects. Furthermore, it is plausible to consider that the frequency of use of the different processes and techniques or project management may be only one of the variables to be considered in order to identify eventual differences among the various types of projects or their complexity. Based on this, a possible issue for new research would be to consider other variables, such as the way or intensity in which the processes and techniques are used by the project managers. Even considering the research limitations, it points out that the scope, people, and communications management are the keys to dealing with complex projects.
The conclusions of this research, considering it is exploratory, not random, and thus, not possible to be generalized, are expected to contribute to the formulation of better constructs suited to the study of complex projects and its management. The conclusions, however, indicated further research is needed on the complexity of projects.
