October 26, 2026
Hybrid model evaluation in project management for urban railway works
Mádava Papalardo Braga; Leticia Franceschini Rodrigues
DOI: 10.22167/2675-6528-202601150
Article prepared by the ResumeAI tool, an artificial intelligence solution developed by the Pecege Institute focused on synthesis and writing.
Summary
This case study evaluated the implementation of a hybrid project management model, which combined traditional governance inspired by PMBOK with flow-oriented execution via Kanban, in the Civil Engineering Management of an urban train operator. The objective was to analyze the impacts of this model on the schedule, predictability, and flow of demands in a portfolio of small and medium-sized urban railway works between 2023 and 2024. A mixed approach was adopted, with quantitative analysis of flow metrics (lead time, throughput, Work in Progress, blockages, and rework) and qualitative evaluation through a Likert scale questionnaire and Net Promoter Score (NPS) with the team and stakeholders. In the comparison of six months before and six months after implementation (Jan-Jun/2023 vs Jan-Jun/2024), a reduction in median lead time from 68 to 43 days (-37%), an increase in average throughput from 16 to 23.5 deliveries/month (+47%), and a decrease in average Work in Progress (WIP) from 38 to 22 items (-42%) were observed. Predictability improved, with on-time deliveries increasing from 52% to 79% and average delay decreasing from 19 to 8 days. The results indicated that the combination of traditional governance with visual flow management increased transparency, reduced interruptions, and improved delivery capacity, even under operational constraints typical of the urban railway environment.
Keywords: Civil engineering; Flow management; Railway infrastructure; Kanban; Hybrid model.
1. Introduction
Project management is strategic for organizations in contexts of high complexity, with multiple stakeholders and pressure for deadlines, quality, and efficiency. The ability to plan, execute, and control initiatives is crucial for delivering value and aligning with institutional objectives. The Project Management Institute (PMI, 2021) defines projects as temporary endeavors for unique results, requiring practices adapted to uncertainty. Kerzner (2017) highlights that success depends on organizational maturity in integrating governance, processes, and people.
In the infrastructure sector, especially in urban railway civil engineering, project management faces additional challenges. Interventions occur during continuous operation, under safety restrictions, limited execution windows, and the need for inter-area compatibility. Planning and executing works in this complex environment amplify the need for coordination and predictability. Carvalho and Rabechini Jr. (2011) observe that, in infrastructure, bureaucracy, interfunctional dependencies, and pressure for results make management susceptible to failures and delays.
Despite the relevance of traditional planning, its isolated application is limited in environments with high operational variability. Rigid models, based on schedules, respond with less agility to changes and blockages (Shenhar and Dvir, 2007). In scenarios of multiple demands and rapid responses, formal control does not guarantee fluidity and predictability. Flow management approaches gain importance, increasing work visibility, identifying bottlenecks, and disciplining the entry of new demands.
The Kanban method, for example, proposes the visualization of the workflow, explicit policies, and limitation of work in progress (WIP), focusing on continuous improvement and predictability (Anderson, 2010). Little’s Law (1961) demonstrates that reducing WIP and stabilizing demand input decreases lead time and increases predictability. Prioritizing flow efficiency over maximizing resource utilization results in faster deliveries and less rework (Reinertsen, 2009; Modig and Åhlström, 2012).
Despite the benefits of flow approaches, their exclusive application may be insufficient in engineering and infrastructure, where activities require clear scope, risk management, and formal change control. Thus, hybrid models, which combine predictive elements (traditional governance, milestones, change control) and adaptive elements (visual flow management, short cadences), have been widely studied (Špundak, 2014; PMI, 2017). This integration is more suitable for scenarios requiring predictability and continuous adaptation, with portfolio-level Kanban (Leopold and Kaltenecker, 2015) and probabilistic metrics (Vacanti, 2015) enhancing management.
In the context of Civil Engineering Management for an urban train operator, the increase in demands and the need for service continuity revealed limitations in the existing management model. The lack of integrated flow visibility, frequent reprioritization, and high volume of work in progress compromised deadline predictability and inter-area coordination. Operational restrictions and the absence of clear rules for emergencies resulted in queues, blockages, and rework. From January to June 2023, the volume of demands received exceeded those completed, with the average lead time ranging between sixty and seventy-five days, evidencing WIP accumulation and bottlenecks.
Faced with this problem, it becomes essential to investigate management alternatives that reconcile control and operational flexibility. The relevance of this study lies in the possibility that improvements in flow management generate significant impacts on deadlines, transparency, prioritization, and delivery capacity in organizations with strong operational pressure, contributing to internal efficiency and service reliability. Thus, this study aimed to evaluate the impacts of adopting a hybrid project management model on civil engineering performance, considering indicators of deadlines, predictability, and delivery capacity.
2. Material and Methods
This study was characterized as an applied research, of a descriptive-exploratory nature, conducted through a mixed approach, which integrated quantitative and qualitative data. The design adopted was that of a single case study, focused on the Civil Engineering Management of a concessionaire responsible for the operation of urban trains in the Metropolitan Region of a large city.
The unit of analysis was the Civil Engineering Management, responsible for small and medium-sized demands, such as renovations and adaptations in platforms, operational buildings, walkways, and bridges. The performance of this management occurs in an active operational environment, with safety restrictions, the need for compatibility with shutdown windows, and supply dependency.
The temporal scope of the study covered the portfolio of demands of this management in the period from January 2023 to December 2024. The main comparison was carried out between two equivalent intervals: January to June 2023, representing the pre-implementation period of the model, and January to June 2024, corresponding to the post-implementation period, when the model was already stabilized.
The quantitative data were obtained from the operational records of the Civil Engineering Management’s demand portfolio, consolidated in spreadsheets and in a visual management board’s monitoring panel. Each analyzed item corresponded to a formally registered demand or work, with the date it entered the system, its movement history between stages, and the completion date.
For analytical standardization purposes, the request date was considered the moment the demand formally entered the portfolio, and the delivery date was considered the moment of final acceptance of the completed demand. The stages observed in the workflow included Backlog, Analysis/Design, Planning/Window, Procurement/Contracting, Execution, Commissioning, and Delivered.
The blockages corresponded to situations where demand remained stalled due to identified and registered impediments, such as absence of an operational window, project pending issues, purchase delays, supplier unavailability, or field interferences. These records were fundamental for the workflow analysis.
In addition to operational records, qualitative data obtained through a structured questionnaire, presented in Appendix I of the original work, were used. The instrument was designed to assess the adoption of the hybrid model in dimensions such as clarity of priorities, visibility of demand status, deadline predictability, scope change management, integration between areas, rework, and overall satisfaction.
The questionnaire consisted of closed-ended questions on a Likert scale from 1 to 5, a Net Promoter Score (NPS) question on a scale from 0 to 10, and open fields for qualitative comments. The questionnaire was administered online during the model consolidation period, asking participants to evaluate the perceived scenario after implementation and a retrospective appreciation of the previous scenario.
The selection of respondents was intentional, including 34 professionals directly involved in the analyzed flow and the main stakeholders who interacted with the Civil Engineering Management. The sample included professionals from the management itself, internal stakeholders from the Operations, Maintenance, Safety, and Supply areas, and representatives of contracted companies.
The study was organized into six main stages: 1) Survey and mapping of the current flow; 2) Diagnosis of bottlenecks and waste; 3) Design of the Kanban system; 4) Implementation and Change Management; 5) Monitoring and refinement; and 6) Evaluation of qualitative and quantitative results.
The construction management process in the pre-implementation period (January to June 2023) was described, mapping the demand input channels, prioritization criteria, roles involved, and points of waiting or rework. This description served as a basis for comparison with the post-implementation scenario.
The hybrid system design planned the combination of a Kanban board with flow steps, explicit policies for readiness and acceptance, work-in-progress (WIP) limits per stage, service classes, and management cadences (daily, weekly prioritization, bi-weekly review, and flow retrospective). Additionally, traditional governance artifacts were incorporated.
Among the traditional artifacts, a simplified Project Charter, a milestone schedule, and a risk matrix were adopted. The Project Charter recorded the identification of the demand, objective, summarized scope, assumptions, constraints, responsible parties, and risks. The milestone schedule organized monitoring points such as approval, design, operational release, execution, and final acceptance.
The risk matrix included categories such as operational safety, window unavailability, supply delays, and field interferences, with indications of impact, response action, and responsible for monitoring. The Change Control Board (CCB) was established to analyze and formalize changes in scope, deadline, cost, and operational impacts, meeting bi-weekly.
The model implementation occurred from July 2023, with the participation of the Civil Engineering Management team and alignment with internal stakeholders. The process was structured in three phases: the first (July to September 2023) focused on organizing the portfolio into a single queue and defining the Kanban board with initial WIP limits.
The second phase (October to December 2023) consolidated management cadences, such as weekly prioritization and review meetings, and structured the functioning of the CCB. The third phase (January to June 2024) involved fine-tuning, with review of WIP limits, refinement of service classes, and expansion of data usage for active impediment management.
Within the scope of Kanban, the flow was organized into columns that represented the process stages: Backlog, Analysis/Design, Planning/Window, Purchasing/Hiring, Execution, Commissioning, and Delivered. This visualization of the flow allowed for tracking the status of demands and identifying bottlenecks and blockages.
To discipline the entry and progression of demands, explicit policies for readiness and acceptance were defined at each stage, as well as service classes to differentiate emergency, regulatory, operational, and improvement demands. WIP limits were established per stage to avoid overload and excessive simultaneous fronts.
Flow and predictability metrics were extracted from the portfolio records. Lead time was defined as the number of calendar days between the formal entry of demand and its final delivery. Throughput corresponded to the quantity of demands completed per month, and WIP was the total number of items simultaneously in progress, located between the Analysis/Design and Commissioning stages.
The accumulated backlog, the percentage of blocked items, rework (measured by reopened demands), and predictability, calculated from the percentage of deliveries completed within the deadline, with a tolerance of up to seven days in relation to the final milestone, were also analyzed. Additionally, the 85th percentile of lead time was observed to capture the behavior of the most delayed items.
Descriptive statistics were applied, including mean, median, and percentiles, and a before/after comparison was performed between the periods of January to June 2023 and January to June 2024. For the blocking analysis, the Pareto principle by category was used. Quantitative data were triangulated with the respondents’ perception, obtained via Likert questionnaire and NPS, and qualitative comments.
The choice for descriptive statistics, without the application of inferential tests, was justified by the exploratory nature of the study and by the nature of the single-case research design, without a control group (Yin, 2018). The triangulation between quantitative and qualitative data strengthened the internal validity of the conclusions (Creswell and Creswell, 2018).
The retrospective survey of “before” perception was recognized as a methodological limitation, which can introduce recall bias. To mitigate this effect, perception data were confronted with independent quantitative evidence, seeking consistency between the two sources of information.
3. Results and Discussion
Project management, in complex organizational contexts, requires the ability to plan, execute, and control initiatives in a structured manner for value delivery and alignment with institutional objectives. The Project Management Institute (PMI, 2021) emphasizes that projects are temporary endeavors to create unique results, demanding practices adapted to uncertainty. Kerzner (2017) complements this, highlighting that success depends on organizational maturity in integrating governance, processes, and people, a challenge particularly accentuated in infrastructure environments.
In the urban railway infrastructure sector, project management faces additional challenges, such as interventions during continuous operation, safety restrictions, and limited execution windows. Carvalho and Rabechini Jr. (2011) point out that, in this environment, factors like bureaucracy, interfunctional dependencies, and pressure for results make management susceptible to failures and delays. Purely traditional management models, based on rigid schedules, demonstrate less agility in responding to changes and blockages, as observed by Shenhar and Dvir (2007).
Given operational variability, flow management approaches, such as Kanban, gain relevance. Kanban, according to Anderson (2010), proposes the visualization of the workflow, explicit policies, and limiting work in progress (WIP), focusing on continuous improvement and predictability. Little’s Law (1961) establishes that reducing WIP and stabilizing demand input decreases lead time and increases predictability, prioritizing flow efficiency over maximizing resource utilization (Reinertsen, 2009; Modig and Åhlström, 2012).
However, the exclusive application of adaptive approaches may be insufficient in engineering and infrastructure, where clear scope definition, risk management, and formal change control are essential. Hybrid models, which combine predictive elements (traditional governance) and adaptive elements (visual flow management), have been studied as more suitable solutions for scenarios requiring predictability and continuous adaptation (Špundak, 2014; PMI, 2017). Leopold and Kaltenecker (2015) and Vacanti (2015) expand this view, proposing portfolio-level Kanban and probabilistic metrics to enhance management.
In the context of the Civil Engineering Management of the urban train operator, the management model prior to the implementation of the hybrid model (January to June 2023) presented significant limitations. Planning was predominantly monthly, based on schedules with little adherence to the variability of urgencies, and reprioritization occurred informally, via emails and direct requests, without a change control process. Monitoring the status of demands depended on spreadsheets and monthly reports, which were frequently outdated.
During this pre-implementation period, the volume of received demands consistently exceeded that of completed demands, and the median lead time varied between sixty and seventy-five days. This situation evidenced an accumulation of work in progress (WIP), with bottlenecks in stages such as design/compatibility, procurement, and operational release. Furthermore, frequent rework was observed, resulting from late scope changes, compromising the predictability and efficiency of deliveries.
Faced with these problems, the Civil Engineering Management opted for the implementation of a hybrid project management model. This model combined traditional governance practices, such as the use of a simplified opening term, milestone schedules, risk matrix, and a Change Control Board (CCB), with visual management and flow control mechanisms inspired by Kanban. The choice of this combination aimed to reconcile the need for formal control of railway works with operational flexibility to deal with urgencies and blockages.
The simplified project charter began to record the demand identification, objective, summarized scope, assumptions, constraints, involved areas, responsible parties, and main risks. The milestone schedule organized the demand monitoring points, such as approval, design, operational release, mobilization, execution, commissioning, and final acceptance. The risk matrix supported the monitoring of critical events, such as window unavailability, supply delays, and field interferences, while the CCB acted as the formal instance for deliberation on scope, deadline, cost, and operational impact changes.
In parallel, the adaptive dimension of the hybrid model was structured with the Kanban flow logic. The work began to be visualized on a single Kanban board, with columns representing the process stages: Backlog, Analysis/Design, Planning/Window, Purchasing/Hiring, Execution, Commissioning, and Delivered. This near real-time visualization allowed for the identification of bottlenecks, the clarification of impediments, and the discipline of new demand entry into the system, according to Anderson’s (2010) principles.
Work in progress (WIP) limits were defined per stage to avoid excessive simultaneous fronts and reduce team overload. Explicit policies for readiness and acceptance in each phase, as well as service classes to differentiate emergency, regulatory, operational, and improvement demands, were established. Management cadences included short alignment meetings to address impediments, weekly prioritization with demanding areas, and bi-weekly reviews with stakeholders, in addition to bi-weekly CCB meetings for formalizing changes.
The model implementation occurred in three phases. The first, between July and September 2023, focused on structuring, with workflow mapping, bottleneck identification, and definition of the Kanban board and initial WIP limits. The second phase, from October to December 2023, consolidated management practices and cadences. The third phase, from January to June 2024, was for stabilization and refinement, with adjustments to WIP limits, improvement of service classes, and expanded use of indicators such as aging and Pareto of blockages for active impediment management.
Qualitative assessment (participant perception)
The qualitative evaluation, conducted through a Likert questionnaire and NPS with 34 participants (managers, engineers, analysts, internal stakeholders, and suppliers), revealed a consistent improvement in the perception of those involved after the implementation of the hybrid model. On the Likert scale of 1 to 5, the clarity of priorities evolved from 2.3 to 4.1, the visibility of demand status from 2.1 to 4.3, and the predictability of deadlines from 2.4 to 3.9. These results indicate that the introduction of the Kanban board, the single entry queue, and regular follow-up cadences made the flow more explicit for the team and stakeholders.
There was also progress in the perception of scope change management, which went from 2.0 to 4.0, and in the reduction of rework, from 2.2 to 4.0. These data suggest that the combination of formal governance and visual management contributed to greater discipline in the flow, fewer occurrences of reopenings, and better coordination among those involved. The improvement in the perception of transparency, visibility, and clarity of priorities is consistent with the Kanban proposal to make work visible and manage flow (Anderson, 2010).
The perception of delivery quality also showed improvement, which can be associated with the clearer definition of done and acceptance criteria, more structured change control, and the reduction of rework throughout execution. The perception of speed and predictability, dimensions directly linked to the system’s ability to balance demand and capacity, also registered gains. This suggests that limiting work in progress, prioritizing discipline, and visibility of blockers increased confidence in meeting deadlines, even in an environment with operational constraints.
In the Net Promoter Score (NPS), the evolution was significant, moving from -12 in the previous scenario to +41 after the implementation of the model. This result was accompanied by an increase in the percentage of promoters from 18% to 56% and a reduction in detractors from 30% to 15%, indicating an expansion of the model’s acceptance among participants and a greater willingness to recommend it to other technical managements. Qualitative comments reinforced this trend, highlighting greater transparency of the workflow, clarity of prioritization rules, ease of identifying impediments, and improved communication between areas.
Despite the improvements, the reports also indicated that some challenges remained relevant, especially those related to the release of operational windows, field interferences, and external dependencies. These elements, which are not eliminated by the adoption of the model, began to be addressed with greater visibility and coordination. Together, the qualitative results suggest that the hybrid model not only influenced operational indicators but also positively altered the perception of those involved regarding portfolio management, which is crucial in railway infrastructure contexts.
Quantitative assessment (flow metrics)
The quantitative evaluation compared operational indicators from January to June 2023 (pre-implementation) with January to June 2024 (post-implementation). A reduction in median lead time from 68 to 43 days was observed, representing a 37% drop. The 85th percentile of lead time, which captures the behavior of the slowest items, also decreased from 128 to 82 days, a reduction of 36%, indicating greater predictability in delivery times.
The average throughput, which measures the quantity of completed demands per month, increased from 16 to 23.5 deliveries, a 47% increment. This increase suggests an expansion of the team’s capacity to complete demands. The reduction in average Work in Progress (WIP) from 38 to 22 items, a 42% drop, indicates less dispersion of efforts and greater focus on finalizing tasks already initiated, aligning with Little’s Law (1961) and Modig and Åhlström’s (2012) principles on flow efficiency.
Process stability indicators also showed improvement. The percentage of blocked items fell from 27% to 14%, and rework due to reopening reduced from 18% to 7%. In terms of predictability, deliveries made within the deadline (with a tolerance of up to seven days) increased from 52% to 79%, a rise of 27 percentage points. The average delay of late deliveries decreased from 19 to 8 days, a reduction of 58%, and the average variance from the baseline reduced from +14 to +4 days, a drop of 71%.
Regarding scope changes, cases without formal approval dropped from 31 to 6, while 18 changes began to be formally handled via CCB. This data indicates a strengthening of process governance and a reduction of “invisible instability” that used to inflate deadlines and generate rework. These quantitative results reinforce the perception that the combination of traditional governance and visual flow management contributed to making the system more stable, predictable, and demand-completion oriented, even in a complex operational environment.
Discussion of results in light of the flow
The reduction of average WIP from 38 to 22 items is consistent with the logic of queueing theory, which postulates that fewer simultaneous items in processing reduce waiting times between stages and decrease “jumping” or context switching in project, purchasing, and execution. This improvement in work fluidity is a direct reflection of the application of WIP limits and explicit Kanban policies, as advocated by Anderson (2010).
The 47% increase in average throughput suggests that the team began finishing more items before starting new ones, improving the completion rate without relying on resource expansion. This observation aligns with the arguments of Reinertsen (2009) and Modig and Åhlström (2012), who emphasize the importance of prioritizing flow efficiency over maximizing resource utilization for faster deliveries and reduced rework. The drop in blockages and rework indicates that explicit policies and short cadences have increased the early detection of impediments and the quality of demand definition before execution.
Implications for operational continuity
The results demonstrate that the hybrid model contributed significantly to the operational continuity of the urban railway. The reduction in lead time, WIP, and blockages, associated with the increase in throughput and on-time deliveries, indicates that the Civil Engineering Management improved its capacity to plan and execute interventions without increasing the level of instability for operational areas. By making the flow more visible and disciplined, the system began to favor more proactive decisions and better coordination among the involved actors, including Operations, Maintenance, and Supplies.
Despite the improvements, the availability of operational windows remains a structural constraint in the urban railway context, not being eliminated solely by changes in the management approach. However, the data indicate that the hybrid model allowed this constraint to be managed with greater predictability, by integrating the logic of baseline and change management with the continuous visibility of the Kanban board. This made it clearer which demands were awaiting release, which depended on supplies or approvals, and which presented a higher risk of delay, reducing surprises for areas such as Operations and Safety.
The observed improvements are not limited to the internal performance of the Civil Engineering Management, but directly impact the continuity of the railway service. By aligning milestones, priorities, and operational windows with greater transparency, the hybrid model favors more informed negotiations, reduces unplanned interferences, and enhances the reliability of the decision-making process for interventions carried out in an active operational environment. The adoption of service classes and WIP limits has proven effective in disciplining the entry of emergency demands without compromising the delivery capacity of planned fronts, a crucial aspect for service stability.
In summary, the implementation of the hybrid project management model in the Civil Engineering Management resulted in substantial improvements in deadline, predictability, and demand flow indicators. The combination of traditional governance with visual flow management via Kanban allowed for a significant reduction in median lead time, an increase in throughput, and a drop in work-in-progress volume, in addition to enhancing the perception of transparency and the satisfaction of those involved. These findings demonstrate that the hybrid model is an effective approach to reconcile control and flexibility in urban railway civil engineering environments, with positive impacts on delivery capacity and operational continuity.
4. Conclusion
This study aimed to evaluate the impacts of adopting a hybrid project management model on civil engineering performance, considering indicators of schedule, predictability, and delivery capability. A significant reduction in median lead time was observed, decreasing from 68 to 43 days, and an increase in average throughput from 16 to 23.5 deliveries per month. A decrease in average Work in Progress was also noted, from 38 to 22 items, indicating a greater focus on completing demands. Predictability improved, with on-time deliveries rising from 52% to 79%, and average delay decreasing from 19 to 8 days. Qualitatively, participant perception revealed advancements in clarity of priorities, visibility of demand status, and schedule predictability, with the Net Promoter Score evolving from -12 to +41. The main contribution lies in demonstrating that the combination of traditional governance, with change control and milestones, and visual flow management via Kanban, with WIP limits and cadences, is an effective approach to reconcile control and flexibility in urban railway civil engineering environments. This resulted in greater transparency, reduced interruptions, and improved delivery capability, positively impacting the operational continuity of the railway service.
However, the study presented limitations, such as the comparison focused on six-month windows, the influence of external factors like the availability of operational windows and supplier performance, and a possible memory bias in the retrospective evaluation of participants’ perception. For future studies, it is suggested to deepen the analysis of service classes and risk policies, automate the collection of flow and aging metrics, and expand the application of the framework to other technical areas, such as electrical and permanent way, evaluating their effects on operational cost and safety. The adoption of a Cumulative Flow Diagram (CFD) with continuous monitoring is also recommended to improve flow management.
Bibliographic References
ANDERSON, David J. Kanban: Successful Evolutionary Change for Your Technology Business. Sequim: Blue Hole Press, 2010.
CARVALHO, Marly M.; RABECHINI JR., Roque. Fundamentos em Gestão de Projetos: Construindo Competências para Gerenciar Projetos. 4. ed. São Paulo: Atlas, 2011.
CRESWELL, John W.; CRESWELL, J. David. Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. 5. ed. Thousand Oaks: Sage, 2018.
KERZNER, Harold. Project Management: A Systems Approach to Planning, Scheduling, and Controlling. 12. ed. Hoboken: Wiley, 2017.
LEOPOLD, Klaus; KALTENECKER, Siegfried. Kanban Change Leadership: Creating a Culture of Continuous Improvement. Hoboken: Wiley, 2015.
LITTLE, John D. C. A Proof for the Queueing Formula L = AW. Operations Research, v. 9, n. 3, p. 383-387, 1961.
MODIG, Niklas; ÅHLSTRÖM, Pär. This is Lean: Resolving the Efficiency Paradox. Stockholm: Rheologica Publishing, 2012.
PROJECT MANAGEMENT INSTITUTE (PMI). A Guide to the Project Management Body of Knowledge (PMBOK Guide) – Seventh Edition. Newtown Square: PMI, 2021.
PROJECT MANAGEMENT INSTITUTE (PMI). Agile Practice Guide. Newtown Square: PMI, 2017.
REINERTSEN, Donald G. The Principles of Product Development Flow: Second Generation Lean Product Development. Redondo Beach: Celeritas Publishing, 2009.
SHENHAR, Aaron J.; DVIR, Dov. Reinventing Project Management: The Diamond Approach to Successful Growth and Innovation. Boston: Harvard Business School Press, 2007.
VACANTI, Daniel. Actionable Agile Metrics for Predictability: An Introduction. Seattle: ActionableAgile Press, 2015.
YIN, Robert K. Case Study Research and Applications: Design and Methods. 6. ed. Thousand Oaks: Sage, 2018.
ŠPUNDAK, Mario. Mixed Agile/Traditional Project Management Methodology – Reality or Illusion? Procedia – Social and Behavioral Sciences, v. 119, p. 939-948, 2014.
Article originating from the Final Course Project of the Specialization in Project Management of the MBA USP/Esalq
To learn more about the course, click here and access the MBX Academy platform