Hyperscale data centers are expanding globally at unprecedented speed, with owners seeking standardised delivery models and identical baseline durations across regions. Yet evidence suggests that local conditions systematically reshape project schedules, even when the design and planned duration remain constant. This paper examines how a 30-month reference baseline for a modular 60 MW data center plays out differently in Spain, the United Arab Emirates, and Singapore.
Findings show that risks are front-loaded in Europe due to permitting and grid connection delays, execution-heavy in the Middle East due to climate and labor constraints, and backloaded in Asia due to strict commissioning and compliance requirements. Drawing on secondary data and case evidence, the study proposes a Global Baseline + Regional Calibration Framework that preserves comparability across regions while accounting for local realities.
The paper contributes to project management theory by extending schedule classification models to include geographic determinants and offers practical guidance for PMOs delivering global data center portfolios.
Data centers are the critical infrastructure of the digital economy, supporting cloud services, streaming, e-commerce, and artificial intelligence. Global IP traffic increased nearly 100-fold between 2005 and 2020 (Cisco, 2020).1
A single hyperscale facility can require more than 100 megawatts (MW) of power, the same as a mid-sized city. These projects can cost over 1 billion USD each. Despite the scale, owners expect delivery within 24–30 months, compared to 4–5 years for industrial plants (CBRE, 2023).2
Mechanical and electrical systems typically account for most data center costs, estimated at around two-thirds of the overall budget. They are delivered on fast-track schedules, where civil works overlap with design and procurement. Redundancy requirements demand rigorous commissioning, often stricter than in nuclear projects. Global operators expect identical standards worldwide, whether in Virginia, Dublin, Dubai, or Singapore (Synergy Research, 2023).3
This raises a key question: If the same data center design is built in different regions, can it really be delivered in the same duration?
The global growth of data centers has been widely documented in industry reports and consultancy insights. Demand is driven by cloud adoption, streaming, and artificial intelligence workloads, resulting in accelerated construction programmes across regions. McKinsey (2023)4 observes that hyperscale operators pursue more cost-efficient and rapid delivery models, even as technical and regulatory challenges increase. Similarly, Turner & Townsend (2023)5 report that costs are rising and supply chain delays for MEP equipment remain a key delivery constraint.
From a project management perspective, data center construction is increasingly characterised as MEP-intensive, fast-tracked, and commissioning-driven. According to Deloitte (2022)6, mechanical and electrical systems typically account for around two-thirds of overall project expenditure. Gray (2023)7 emphasises the shortening of delivery timelines despite added technical requirements for cooling and redundancy. Plan Radar (2024)8 highlights that compliance with efficiency and regulatory standards has become equally important as meeting physical delivery milestones.
The project scheduling literature provides a foundation for understanding how such complexities interact with planning. Classical approaches, including CPM and resource-constrained scheduling, are well-established in construction contexts (IJETT, 2023)9. Recent research highlights the integration of digital tools such as BIM and AI to generate more adaptive schedules. Yet, these studies typically address generic construction sectors—residential, commercial, or infrastructure—and rarely focus on data centers.
Regulatory and permitting frameworks also play a significant role. The World Bank’s Doing Business data reveals regional disparities in permitting timelines (e.g., 147 days for Spain) (World Bank, 2020),10 while European legislation mandates Environmental Impact Assessments for large-scale projects (European Commission, 2014).11 In the U.S., county-level zoning studies highlight how local rules drive delivery cycles. In Singapore, new capacity allocations are linked to energy-efficiency targets such as Power Usage Effectiveness, PUE ≤ 1.3, effectively embedding sustainability into the approval process (IMDA, 2022).12
Despite this growing body of work, no scholarly article has directly examined how identical data center projects unfold across multiple regions under the same baseline duration. This study addresses that gap by analysing how an identical 30-month baseline data center project plays out in Spain, the UAE, and Singapore.
This study employs a comparative case study design, focusing on the application of an identical 30-month baseline schedule for a 60 MW modular hyperscale data center in three distinct regional contexts: Spain, the United Arab Emirates (UAE), and Singapore.
The analysis draws primarily on secondary data, integrating regulatory benchmarks, industry indices, and case evidence from the 2020–2024 period. Sources include the World Bank’s Doing Business database for permitting timelines, European Commission directives for environmental assessments, UAE labor regulations such as the midday work ban, and Singapore’s Green Mark scheme for data centers. Industry perspectives were incorporated from Turner & Townsend’s Data Centre Cost Index, CBRE’s global trends reports, Uptime Institute surveys, and press releases from developers and regulators.
To ensure comparability, project activities were coded against five work breakdown structure (WBS) categories common to data center delivery: design and permitting, procurement, civil and structural works, MEP installation, and commissioning and certification. The evidence was then pattern-matched against the baseline to identify where risks and delays concentrated in each geography. While this approach provides valuable insights into how identical schedules unfold under different conditions, it is constrained by reliance on published data rather than direct access to live project schedules. As such, the framework developed here emphasises regional risk patterns rather than precise activity durations, with validation through primary project data left for future work.
The comparative analysis assessed where risks concentrate along the baseline schedule in each geography. A pattern-matching technique was used to align empirical evidence with the baseline model.
Large projects face 147 days on average just for building permits (World Bank, 2020). Most must undergo a full Environmental Impact Assessment, adding months to the timeline. Grid access is another bottleneck: in 2023, Spanish developers reported major delays connecting to the national grid..
Permitting is faster in the UAE, but climate and labor laws dominate execution. From June to September, outdoor work is banned from 12:30–15:00 daily, reducing productivity. Average highs of 45°C make night shifts and special concreting methods essential. The workforce is over 80% expatriate, shaping training and QA/QC strategies (United Nations, 2022).13
Since 2022, project approvals in Singapore have been conditional on compliance with green allocation standards, which has led to extended commissioning phases of around 20–30%, including additional testing and certification (Uptime Institute, 2022).14
The findings demonstrate that although the projects share an identical baseline, regional conditions cause the critical path to diverge, underscoring that context is not merely a risk factor but a structural determinant of schedule feasibility. This insight aligns with PMI’s PMBOK Guide emphasis on tailoring, while also challenging AACE/DCMA schedule classifications to more fully account for regional determinants.
Figure 1 illustrates how the identical 30-month baseline is reshaped across Spain, the UAE, and Singapore, making visible how the point of schedule pressure migrates depending on the regional context.
Here, the criticality shifts to the final stages of testing and validation, making commissioning the new critical path.
The broader insight is that while the baseline duration remains fixed, the location of risk shifts: front-loaded in Spain, execution-phase in the UAE, and backloaded in Singapore. This demonstrates that regional conditions do not simply add delays, but reconfigure the shape of the schedule itself, with direct implications for where PMOs must allocate contingency and management attention.
The findings suggest that a paradox lies at the heart of global data center delivery: operators seek comparability across projects, yet local realities consistently reshape the critical path. To address this, the study advances a Global Baseline + Regional Calibration Framework. At its foundation, the framework begins with a global baseline—an agreed reference WBS and duration that allows portfolio-level benchmarking, in this case, a 30-month delivery model. This baseline provides the anchor against which regional variations can be assessed.
The second step introduces regional calibration, which acknowledges that the same baseline cannot be achieved uniformly across contexts. Evidence indicates that in Spain, permitting and grid connection issues justify a six-month extension; in the UAE, climate constraints and labor productivity require a three-month buffer during execution; and in Singapore, stringent commissioning and compliance obligations add approximately four months to the back end of delivery. By incorporating these adjustments, global PMOs preserve the comparability of their schedules while realistically accounting for local determinants.
Finally, the framework emphasises portfolio-level synthesis. Applying regional calibrations enables project schedules to be compared directly without obscuring the impact of external context. In doing so, it provides PMOs with a structured means of reconciling global standardisation with local realities, while enhancing transparency for sponsors and improving the reliability of delivery forecasts. A visual representation, such as a comparative timeline diagram, would further illustrate how an identical design expands differently across regions, making these dynamics more accessible to both practitioners and researchers.
The study contributes to both theory and practice in project management. Theoretically, it extends schedule classification models, such as those advanced by AACE and DCMA. In doing so, it advances the tailoring principle outlined in PMI’s PMBOK Guide by offering a quantifiable framework through which tailoring can be operationalised in global infrastructure scheduling. The research also addresses a gap in the literature by focusing specifically on data center delivery, a sector that has received limited attention in scholarly project management research despite its rapid global expansion.
On the practical side, the framework offers PMOs a replicable method for calibrating schedules across their portfolios. It offers guidance on how regional determinants such as permitting, climate, labor, and commissioning can be incorporated into schedules while still preserving the comparability required by global operators. Importantly, the study also highlights the need to elevate commissioning to a distinct phase in data center schedules, reflecting industry evidence that testing and certification have become decisive bottlenecks.
Collectively, these contributions situate the study at the intersection of theoretical advancement and applied project management practice, bridging conceptual models with real-world scheduling challenges.
Like any comparative study, the analysis presented here has limitations. The reliance on secondary data means that certain dynamics—such as unrecorded delays or contractual strategies—may be under- or over-represented. The focus on three regions, while sufficient to illustrate the framework, does not capture the diversity of delivery conditions found in other markets, particularly in emerging economies. Moreover, the framework does not directly account for contracting strategies, supply chain models, or governance structures, each of which may substantially influence outcomes.
Future research should therefore seek to test the framework quantitatively using actual baseline and as-built schedules from multiple projects and geographies. Simulation approaches, including Monte Carlo analysis, could be employed to assess how regional calibration affects risk exposure and contingency planning. Additional attention could be given to how contract type—EPC versus design-build, for example, modulates the impact of regional determinants. Finally, the development of a decision-support tool that allows PMOs to dynamically apply regional calibration factors would provide a practical extension of the conceptual model presented here.
This study explored whether an identical data center design, planned for the same duration, can be delivered uniformly across regions. The evidence from Spain, the UAE, and Singapore demonstrates that the answer is no: although each project begins with a 30-month baseline, the critical path diverges in predictable ways. In Spain, risk is concentrated at the front end in permitting and grid access; in the UAE, it shifts into the execution phase due to climate and labor constraints; and in Singapore, it is backloaded into commissioning, driven by stringent sustainability and compliance standards.
These findings confirm that regional context is not a peripheral risk but a structural determinant of schedule feasibility. To address this, the study proposed a Global Baseline + Regional Calibration Framework, which allows global PMOs to preserve comparability while realistically incorporating local determinants. For practitioners, the framework offers a replicable method to calibrate schedules across portfolios; for scholars, it extends schedule classification theory to account for geography as a shaping variable.
One design, one baseline, yet three different realities—this paradox highlights both the promise and the challenge of delivering hyperscale infrastructure in a globalised but locally constrained world.
References:
All Courses
Upcoming Course Dates
Professional Courses
Academy & Short Courses
IPM Courses
PMI Courses
IPMA Courses
PRINCE2 Courses
Corporate Training
Corporate Partnership
Strategic Partnership
IPM Competency Pathway
Certification Overview
IPMA Certification
PMI Certification
About IPM Hub
IPM Membership
EMEA Monthly Events
Mentorship
All Resources
Expert Insights
Ebooks & Toolkits
Publish Your Writing
IPM Jobs Portal
The Institute
Faculty
Global Advisory Board
Apply to Lecture or Volunteer
Apply to Intern or Campus Ambassador
Speak to an Advisor
