Investigating Maritime Ports' Operational Performance Determinants: An Empirical Study of Operational Delays and Moroccan Container Terminal Productivity.

ISSN: 2658-8455 Volume 6, Issue 13 (2025), pp. 324-341. © Authors: CC BY-NC-ND . 324 www.ijafame.org Investigating Maritime Ports’ Operational Performance Determinants: An Empirical Study of Operational Delays and Moroccan Container Terminal Productivity Analyse des déterminants de la performance opérationnelle des ports maritimes : Une étude empirique des retards opérationnels et de la productivité des terminaux à conteneurs marocains Zakaria ELKHARMALI, (Ph.D. Candidate in Economics and Management) Research Team: Management and Information System National School of Business and Management, Tangier Abdelmalek Essaadi University, Tetouan, Morocco Ouail EL KHARRAZ, (Ph. Professor in Economics and Management) Research Team: Management and Information System National School of Business and Management, Tangier Abdelmalek Essaadi University, Tetouan, Morocco Correspondence address : ENCG : Ecole Nationale de Commerce et de Gestion_Tanger. Q553+5JX, Tangier 90000. Abdelmalek Essaadi University Tetouan, Morocco. Phone: 05393-13487 Disclosure Statement : The authors declare that they have not received any financial support that could have influenced the objectivity of this study. They take full responsibility for any potential plagiarism, the use of artificial intelligence in the writing process, as well as for the results presented in this article. Conflict of Interest : The authors report no conflicts of interest. Cite this article : ELKHARMALI, Z., & EL KHARRAZ, O. (2025). Investigating Maritime Ports’ Operational Performance Determinants: An Empirical Study of Operational Delays and Moroccan Container Terminal Productivity. International Journal of Accounting, Finance, Auditing, Management and Economics, 6(13), 324–341. https://doi.org/10.5281/zenodo.17917271 License This is an open access article under the CC BY-NC-ND license International Journal of Accounting, Finance, Auditing, Management and Economics - IJAFAME ISSN: 2658-8455 Volume 6, Issue 13 (2025) Zakaria ELKHARMALI & Ouail EL KHARRAZ. Investigating Maritime Ports’ Operational Performance Determinants: An Empirical Study of Operational Delays and Moroccan Container Terminal Productivity 325 www.ijafame.org Investigating Maritime Ports’ Operational Performance Determinants: An Empirical Study of Operational Delays and Moroccan Container Terminal Productivity Abstract: This study investigates how operational delays influence container terminal productivity by quantifying their effect on Gross Crane Productivity (GCP) at Moroccan container terminals. Based on six months of calls data (April– September 2025), we used a simple linear regression based on a sample of 422 observations.The distribution of GCP shows a slight right skew, with a stable operating range of 28–35 moves per crane-hour, whereas delays are heavily right-skewed, mostly under 10 minutes with occasional long outliers. Ordinary Least Squares results reveal a significant negative association between delays and GCP (β = −0.46, p < 0.001), indicating that each additional minute of delay lowers productivity by about 0.46 moves per hour, delays account for roughly 20% of the variance in GCP (R² = 0.20). LOWESS analysis highlights a tolerance threshold near 10 minutes, after which productivity drops sharply due to congestion propagation and loss of synchronization between quay and yard operations. Over the study period, cumulative delays decreased by nearly half while average GCP improved by approximately 25%, suggesting learning effects, better planning, and enhanced coordination. Overall, the findings show how microlevel time losses affect terminal productivity and connect directly to the smart-port agenda, emphasizing that automation and digital connectivity yield the greatest benefits when supported by integrated scheduling and strong organizational readiness. The study recommends the development of predictive delay analytics (e.g., AIS combined with machine learning) and real-time, integrated equipment scheduling as key levers to sustain both operational efficiency and environmental performance. Mots clés : Operational delays, Gross Crane Productivity (GCP), Automated container terminals, Integrated scheduling, Smart port performance. JEL Classification : R41, L91, C21, C51 Type du papier : Empirical Research Résumé : Cette étude analyse l’influence des retards opérationnels sur la productivité des terminaux à conteneurs en quantifiant leur effet sur la Gross Crane Productivity (GCP) dans les terminaux marocains. À partir de six mois de données d’escales (avril–septembre 2025), nous avons appliqué une régression linéaire simple sur un échantillon de 422 observations. La distribution de la GCP présente une légère asymétrie à droite, avec une plage opérationnelle stable de 28 à 35 mouvements par grue et par heure, tandis que les retards sont fortement asymétriques, majoritairement inférieurs à 10 minutes avec quelques valeurs extrêmes prolongées. Les résultats de l’estimation par moindres carrés ordinaires montrent une association négative significative entre les retards et la GCP (β = −0.46, p < 0.001), indiquant que chaque minute de retard supplémentaire réduit la productivité d’environ 0,46 mouvement par heure ; les retards expliquent environ 20 % de la variance de la GCP (R² = 0.20). L’analyse LOWESS met en évidence un seuil de tolérance autour de 10 minutes, au-delà duquel la productivité chute fortement en raison de la propagation de la congestion et de la perte de synchronisation entre les opérations quai et parc. Sur la période étudiée, les retards cumulés ont diminué de près de moitié tandis que la GCP moyenne s’est améliorée d’environ 25 %, ce qui suggère des effets d’apprentissage, une meilleure planification et une coordination renforcée. Globalement, les résultats montrent comment les pertes de temps au niveau microopérationnel affectent la productivité terminale et s’inscrivent directement dans l’agenda du smart port, soulignant que l’automatisation et la connectivité numérique offrent leurs meilleurs bénéfices lorsqu’elles sont accompagnées d’une planification intégrée et d’une forte préparation organisationnelle. L’étude recommande enfin le développement d’outils prédictifs d’analyse des retards (par exemple, AIS combiné à des techniques d’apprentissage automatique) ainsi que la mise en place d’une programmation intégrée et en temps réel des équipements comme leviers essentiels pour soutenir à la fois l’efficacité opérationnelle et la performance environnementale. Keywords : Retards opérationnels, Productivité Brute des Grues (GCP), Terminaux à conteneurs automatisés, Planification intégrée, Performance smart port. Classification JEL : R41, L91, C21, C51 Paper type : Recherche empirique ISSN: 2658-8455 Volume 6, Issue 13 (2025), pp. 324-341. © Authors: CC BY-NC-ND . 326 www.ijafame.org 1. Introduction As per (UNCTAD, 2021), The maritime sector constitutes the backbone of global trade, ensuring the transport of more than 80% of international goods by volume. It should be noted that the efficiency and performance of ports, as crucial logistical nodes, are essential determinants of economic success at both national and global levels (Haralambides, 2017 and Zhen, 2015). (Lee & Song, 2023; Mishrif & Khan, 2023; Song et al., 2024) state that in the face of major post-pandemic challenges including COVID-19, digitalization, and decarbonization (CDD) the industry is experiencing unprecedented complexity that demands a radical transformation of port operations. In order to address these challenges, the development of Automated Container Terminals (ACTs) has become a priority. Automation, one of the three fundamental pillars of automated container terminals, aims to enhance operational efficiency by reducing human intervention and minimizing congestion and turnaround time (Yen et al., 2023). This evolution has led to an exponential growth in research publications since 2007 (Pallis et al., 2010, cited in Kishore et al., 2024) and to the emergence of a solid body of knowledge on operational optimization. As per (Naeem et al., 2023), automation introduces new complexities, especially the critical issue of Integrated Scheduling of handling equipment: Quay Cranes (QCs), Automated Vehicles (AVs), and Yard Cranes (YCs). Operational success in automated container terminals depends on avoiding interferences among these pieces of equipment (Naeem et al., 2023), as any disturbance in their coordination generates operational delays. These delays often stem from complex factors frequently neglected in theoretical models, such as insufficient buffer capacity beneath QCs or challenges in allocating yard storage locations for import containers (Naeem et al., 2023). (Naeem et al., 2023) state that the impact of these delays directly affects the terminal’s most critical productivity metric such as Gross Crane Productivity (GCP) which serves as the main indicator of efficiency at the ship-to-shore interface. Minimizing delays is therefore essential, as research recommends that scheduling objectives focus on minimizing equipment waiting time, an approach that reduces the vessel makespan while maximizing asset utilization. Despite the centrality of GCP and operational delays, the field of smart port performance assessment is often described as fragmented and primitive, with insufficient research linking theory to practice (Molavi, 2020, cited in Paraskevas et al., 2024; Paraskevas et al., 2024). A major limitation of existing studies lies in the absence of an objective and scientific assessment framework capable of accurately identifying performance problems (Paraskevas et al., 2024), as analyses are often based on theoretical models or heterogeneous datasets. Our study aims to fill this empirical and methodological research gap. The empirical contribution lies in conducting an in-depth analysis based on six months of operational data (April–September 2025) from Moroccan container terminals, providing an objective quantification of productivity losses in an emerging economy where studies are scarce (Alamoush et al., 2020, Paraskevas et al., 2024). The methodological contribution is the combination of exploratory data analysis (EDA), LOWESS smoothing, and OLS regression to rigorously quantify the effect of operational delays. The managerial contribution is the provision of actionable insights and recommendations for terminal planning and real-time operational monitoring, highlighting how delays can be mitigated to improve GCP. Furthermore, the study contributes to the literature by focusing on an emerging economy (Moroccan context) an area where research remains scarce, as most studies focus on ports in Zakaria ELKHARMALI & Ouail EL KHARRAZ. Investigating Maritime Ports’ Operational Performance Determinants: An Empirical Study of Operational Delays and Moroccan Container Terminal Productivity 327 www.ijafame.org developed countries (Alamoush et al., 2020, cited in Paraskevas et al., 2024). The structure of this article is as follows. Section 2 presents the Theoretical Framework, outlining the key concepts of port performance, GCP, and operational delays. Section 3 details the Research Methodology, including the nature of the operational data and the analytical techniques employed. Section 4 presents our Empirical Results, quantifying the impact of operational delays on GCP. Section 5 discusses the theoretical and managerial implications, and Section 6 concludes the article by identifying limitations and suggesting directions for the future research. 2. Literature review 2.1. Port Performance and Operational Efficiency The maritime sector, responsible for transporting over 80% of global trade volume (UNCTAD, 2021), grants ports a crucial role as pillars of the global economy. Port performance efficiency is considered a fundamental determinant of success, directly linked to infrastructure design and the operational environment. Port performance evaluation has experienced exponential growth, with a notable emphasis on quantifying parameters since 2007 (Kishore et al., 2024). Traditional performance indicators, initially focused on internal operational efficiency, are evolving toward a broader approach integrating customer demand, reliability, and service quality (Brooks & Cullinane, 2006, cited in Hardianto et al., 2023). With the rise of containerized trade volumes, Automated Container Terminals (ACTs) have drawn increasing research attention due to their potential to enhance terminal productivity and operational performance. Ports are increasingly urged to become Smart Ports by investing in technologies that promote more efficient operations via Information and Communication Technologies (ICT) and the Internet of Things (IoT). Efficiency as a holistic concept: Modern port performance extends beyond mere technical efficiency. It is a multidimensional concept that must be assessed not only through economic indicators but also through social, environmental, and sustainability criteria. Recent research trends show a strong focus on sustainability, which has become the most relevant topic over the last five years (Kishore et al., 2024). Evaluation methodology: Objective evaluation of port efficiency often relies on Data Envelopment Analysis (DEA), widely used in maritime transport due to its advantage as a nonparametric assessment providing objective measurements (Dewita, Yen, & Burke, 2018, cited in Yen et al., 2023). The DEA-Tobit model is frequently applied to assess the influence of external factors, such as smart port dimensions (Automation, Environment, Intelligence), on measured operational efficiency. Technical efficiency is often evaluated via resource utilization, fleet size, or fuel consumption to produce outputs such as throughput and revenue. Performance evaluation is thus essential for effective management, with managers relying on it for planning and resource optimization. 2.2. Operational Delays as a Determinant of Performance Operational delays are a major source of efficiency loss, additional costs, and customer dissatisfaction in container terminals. The main objective of optimization studies is to minimize these delays. In Automated Container Terminals (ACTs), where all equipment is automatically controlled, avoiding interference among each equipment pair is crucial to prevent disruption or cargo accumulation (backlog). Delays are a direct consequence of coordination failures. Literature highlights several interference and delay factors arising from the complexity of integrated scheduling among three main equipment types: Quay Cranes (QCs), Yard Cranes (YCs), and ISSN: 2658-8455 Volume 6, Issue 13 (2025), pp. 324-341. © Authors: CC BY-NC-ND . 328 www.ijafame.org Automated Vehicles (AVs) (Naeem et al., 2023). Integrated scheduling remains a continuous and complex research domain (Naeem et al., 2023). However, most studies have focused on scheduling a single equipment type, mainly AVs, assuming that scheduling of other equipment (e.g., QCs) is predetermined. Such simplifications render models inadequate when addressing real-world delays arising from complex interdependencies. Neglected delay factors: Storage allocation: The storage of import containers plays a crucial role in YC scheduling. Neglecting this factor affects operational time for both YCs and AVs. Incorporating storage allocation into handling equipment scheduling can yield more efficient solutions (Naeem et al., 2023). Buffer capacity (BC): Buffer capacity beneath QCs is a major constraint that may generate interference, yet it is often ignored in integrated equipment scheduling studies. Performance analyses under different BC configurations provide critical insights for terminal performance assessment (Hu et al., 2013; Xinyan et al., 2014). Tobit modeling: To evaluate the influence of external factors such as environmental conditions or smart port characteristics on efficiency, the Tobit regression model is used. Within this framework, operational delays or their determinants (e.g., level of digitalization, automation status, control systems) can be treated as independent variables to assess their impact on performance. Thus, the literature justifies the causal link: interferences → operational delays → reduction in GCP, highlighting the need for empirical validation in emerging contexts, such as Moroccan terminals. 2.3. Gross Crane Productivity as a Key Performance Indicator Gross Crane Productivity (GCP), measured in crane moves per hour, is one of the most direct and relevant terminal performance metrics, focusing on efficiency at the ship-to-shore interface. GCP is intrinsically linked to the fundamental management goal: minimizing the total time a vessel spends at berth. Minimizing makespan: The most common scheduling objective is to reduce vessel makespan, as it represents a critical performance measure directly influencing vessel berthing time (Luo & Wu, 2020b, cited in Naeem et al., 2023). Equipment-related KPIs: Equipment objectives, such as minimizing QC delay or AV waiting time, have received considerable attention (Yue et al., 2021). Monitoring GCP enables managers to identify bottlenecks and optimize resource allocation, providing actionable insights for integrated planning and performance improvement (Hardianto et al., 2023; Al-Fatlawi & Motlak, 2023). 2.4. Operational Delays and Productivity The relationship between operational delays and GCP is direct: effective delay management leads to immediate gains in productivity and asset utilization. In an automated terminal, any delay in equipment (e.g., an AV) can create a bottleneck, causing crane idleness and instant reduction in GCP. Studies on buffer capacity and storage allocation show how terminal configurations influence GCP under real-world constraints (Hu et al., 2013; Xinyan et al., 2014; Roy et al., 2020). Collecting six months of operational microdata in Moroccan terminals enables quantification of delays and their direct impact on GCP, bridging theoretical scheduling challenges (interference, buffer capacity) with actual productivity outcomes and providing a strong empirical basis for recommendations on integrated scheduling and smart port practices. Based on the above literature review, the research hypotheses are formulated as follows: Zakaria ELKHARMALI & Ouail EL KHARRAZ. Investigating Maritime Ports’ Operational Performance Determinants: An Empirical Study of Operational Delays and Moroccan Container Terminal Productivity 329 www.ijafame.org H1: Operational delays have a negative impact on Gross Crane Productivity (GCP) in container terminals. H2: Structural and technological conditions (e.g., automation, buffer capacity, storage allocation) moderate the relationship between operational delays and GCP. 3. Methodology 3.1. Research Approach and Study Design This research adopts a quantitative, empirical methodology grounded in the analysis of largescale operational data. The aim is to provide an objective and scientific assessment of port efficiency, in contrast to prior studies relying on qualitative and subjective methods, which complicate comparability (Molavi et al., 2020). The study draws on high-fidelity data collected from Moroccan container terminals over a consecutive six-month period (April–September 2025). This approach aligns with research trends prioritizing Big Data and predictive tools for resource management and operational efficiency (Dalaklis et al., 2023, Paternina-Arboleda et al., 2023). Data and Source: The dataset comprises 422 ship calls during the six-month period, covering three shifts per day: morning (06:00–14:00), afternoon (14:00–22:00), and night (22:00–06:00). The data were obtained directly from Tanger Med 2 container terminal operational logs. Unit of Analysis and Data Structure: Each observation corresponds to a single ship call per shift, with data aggregated at the call level. For each observation, operational variables include gross crane productivity (GCP), operational delays, and associated structural/technological factors. Selection Criteria and Data Quality: Observations were included if they corresponded to standard handling operations at the terminal. Extreme or missing values were identified during Exploratory Data Analysis (EDA) and treated according to standard cleaning procedures, including removal or imputation, ensuring the robustness of subsequent regression models. Study Design Justification: The dataset allows testing causal relationships at the micro (call) level, bridging theoretical models of interference, buffer capacity, and storage allocation with actual terminal productivity outcomes. 3.2. Definition and Operationalization of Variables The study employs a causal modeling approach. Dependent Variable: Gross Crane Productivity (GCP), measured in moves per STS-crane hour (moves/hour/crane). GCP reflects vessel berthing efficiency and overall terminal performance (Luo & Wu, 2020b; Naeem et al., 2023). Independent Variable: Operational Delays, measured in minutes or hours, reflecting ship waiting times at berth or anchor. Delays are considered the empirical manifestation of coordination and interference issues (Naeem et al., 2023). Moderating Variables: Structural and technological conditions (automation level, buffer capacity, storage allocation) are included as potential moderators of the relationship between delays and GCP. 3.3. Statistical Analysis and Analytical Framework The methodology combines Exploratory Data Analysis (EDA) and inferential modeling using Python-based tools (Plotly, Pandas, Statsmodels, Seaborn). Phase I: Exploratory Data Analysis (EDA). • Information Exploration: Data cleaning, transformation, and consistency checks ensure suitability for analysis, as required for DEA-type analyses (Hsu et al., 2023). ISSN: 2658-8455 Volume 6, Issue 13 (2025), pp. 324-341. © Authors: CC BY-NC-ND . 330 www.ijafame.org • Distribution Analysis: Kernel density estimates, histograms, and pairplots visualize dependent and independent variable distributions, highlighting outliers. • Spatio-Temporal Analysis: Boxplots and line charts examine GCP and operational delays across months and shifts, identifying trends or seasonal effects. Phase II: Inferential Analysis and Modeling • Equation of Research: The basic regression model can be expressed as: 𝑮𝑪𝑷𝒊 = 𝜷𝟎 +𝜷𝟏𝑫𝒆𝒍𝒂𝒚𝒊 + 𝜷𝟐𝑴𝒐𝒅𝒆𝒓𝒂𝒕𝒐𝒓𝒊 + 𝝐𝒊 Where 𝒊 is the ship call observation, 𝑫𝒆𝒍𝒂𝒚 represents operational delays, and 𝑴𝒐𝒅𝒆𝒓𝒂𝒕𝒐𝒓 represents structural/technological conditions. Correlation Analysis: Spearman and Pearson correlations quantify linear and non-linear relationships between operational delays and GCP. Regression Modeling: • OLS Regression: Used as the primary model to estimate the average effect of operational delays on GCP. Although simple and univariate, it allows interpretation of direct relationships. • Quantile Regression: Employed to examine the effect of delays across the GCP distribution, addressing sensitivity to extreme values and differential impacts on highvs. low-performing calls. Justification of Approach: While the OLS univariate model provides a straightforward baseline, further robustness checks (quantile regression, sensitivity to extreme values) complement it and provide insights into heterogeneous effects across observations. The combination of EDA, correlation, OLS, and quantile regression ensures methodological rigor while maintaining transparency. 4. Results and Analysis This section details the findings of the empirical analysis conducted on six months of operational data (April–September 2025) from Moroccan container terminals. The results are presented in three subsections covering performance trends, correlation assessment, and modeling of the causal impact of operational delays on Gross Crane Productivity (GCP). 4.1. Descriptive Analysis of Variables Fig 1 present an Hitogram of GCP: Figure 1. Histogram of GCP Source: By our sides during data analysis, 2025. Figure 1 shows the empirical distribution of Gross Crane Productivity (GCP) observed at the container terminal over the analysis period. The density curve overlaid on the histogram reveals Zakaria ELKHARMALI & Ouail EL KHARRAZ. Investigating Maritime Ports’ Operational Performance Determinants: An Empirical Study of Operational Delays and Moroccan Container Terminal Productivity 331 www.ijafame.org slight right skewness (positive skewness), indicating that most observations are concentrated around a central interval between 28 and 35 moves per crane-hour. This band represents the average performance level of quay cranes, reflecting sustained yet stable activity under normal operating conditions. A few extreme values appear beyond 50 moves/hour, likely corresponding to exceptional highproductivity situations associated with favorable conditions such as optimal planning, low quay congestion, or a highly efficient handling team. Conversely, the low frequency of observations below 25 moves/hour reflects temporary slowdowns, often attributable to technical interruptions, vessel rotation, or adverse weather. Overall, this distribution indicates moderate variability in crane performance typical of an automated or semi-automated terminal where operations are generally stable but sensitive to certain external factors. The descriptive analysis thus highlights controlled dispersion around the mean, confirming operational consistency at the terminal and providing a basis for analyzing the relationship between operational delays and crane productivity. Fig 2 present an Histogram of Operational Delays: Figure 2. Histogram of Operational Delays Source: By our sides during data analysis, 2025. Figure 2 presents the statistical distribution of operational delays over the study period. The density curve associated with the histogram reveals pronounced right skewness, showing a high concentration of observations at low delay values, mainly between 0 and 10 minutes. This pattern indicates that most port operations are completed within controlled timeframes, reflecting overall operational efficiency at the terminal. However, the long right tail (delays exceeding 20–30 minutes) highlights occasional episodes of inefficiency likely associated with exceptional events: crane incidents, yard saturation, prolonged truck queues, or coordination constraints between quay and yard teams. Although rare, these extreme delays exert a disproportionate effect on overall performance especially on GCP by disrupting normal operational flow. The strong dispersion observed in the upper part of the distribution also indicates non-negligible operational variability, suggesting that certain time windows or vessels face particular difficulties in maintaining the expected cadence. This descriptive analysis therefore reveals the coexistence of predominantly stable performance with localized peaks of inefficiency a key lever for improving cycle time management and optimizing port planning. 4.2. Monthly Performance Trends Fig 3 present a Boxplot of GCP by month: ISSN: 2658-8455 Volume 6, Issue 13 (2025), pp. 324-341. © Authors: CC BY-NC-ND . 332 www.ijafame.org Figure 3. Boxplot of GCP by Month Source: By our sides during data analysis, 2025. Figure 3 displays monthly variation in GCP over the analysis period using boxplots. This visualization enables assessment of dispersion, the median, and potential outliers of GCP for each month from April to September. Overall, there is a gradual upward trend in the monthly median GCP, indicating steady improvement in operational performance. This progression can be explained by several factors: • Adjustments to work schedules and optimization of handling sequences after the initial weeks of observation. • The learning curve of operators and technical teams, leading to better coordination among quay cranes, yard operations, and tractors. • More favorable seasonal conditions in summer, often associated with increased fluidity in maritime operations. Outliers observed in some months correspond to isolated instances of exceptional performance or, conversely, unexpected slowdowns. These isolated points reflect the dynamic and stochastic nature of port operations, where productivity varies with vessel size, handled volume, equipment availability, or stowage plan complexity. Hence, the relatively stable dispersion and the progressive rise in the median suggest operational maturation at the terminal over the study period, trending toward greater stability in crane performance. This evolution confirms the effectiveness of internal measures adopted to improve productivity and mitigate operational contingencies. Fig 4 present the monthly average of GCP: Figure 4. Monthly Average GCP Source: By our sides during data analysis, 2025. Figure 4 confirms the ascending trend seen in Figure 3. It depicts the evolution of average monthly GCP between April and September, during which GCP increases from about 28 Zakaria ELKHARMALI & Ouail EL KHARRAZ. Investigating Maritime Ports’ Operational Performance Determinants: An Empirical Study of Operational Delays and Moroccan Container Terminal Productivity 339 www.ijafame.org From a theoretical standpoint, this research bridges the gap between micro-level operational dynamics and macro-level port performance frameworks. It provides empirical confirmation that time efficiency acts as a key mediator linking automation capacity with overall productivity, extending the perspectives offered by Peng et al. (2022) and Yen et al. (2023). Future studies should expand this model to include multiple terminals and examine how factors such as infrastructure quality, intelligent scheduling algorithms, and human–machine collaboration influence productivity. The use of advanced forecasting techniques such as reinforcement learning or hybrid deeplearning models could further refine decision-support systems for next-generation smart ports, strengthening both operational reliability and sustainability outcomes. 6. Conclusion This study provides clear, data-driven evidence that time losses are a primary constraint on quayside productivity. In the Moroccan terminals analyzed, every minute of operational delay measurably depresses GCP, with a nonlinear drop once delays exceed approximately 10 minutes. Beyond the headline coefficient (β = −0.46), two insights are particularly relevant for practice: • First, delay management is not only a technical matter but an organizational one, as coordination quality among quay cranes, yard cranes, and transport vehicles determines how efficiently automation translates into throughput. • Second, predictive and integrated control is decisive, since combining early-warning delay analytics with dynamic, real-time scheduling can prevent desynchronization, shorten cycle times, and reduce idle energy use. The results also confirm that GCP sits at the intersection of the smart-port performance triad automation, connectivity, and sustainability. As delays declined during the study window, GCP improved accordingly, illustrating how digital maturity and continuous process learning compound efficiency gains. However, the analysis also shows that a substantial share of GCP variance remains unexplained by delays alone, highlighting the influence of factors such as infrastructure capacity, vessel characteristics, equipment reliability, and workforce skill sets. This introduces clear limitations. The study relies on a single-terminal dataset and focuses primarily on linear delay effects, which restricts generalizability and may overlook important threshold dynamics and heterogeneous impacts across operational conditions. Additionally, the use of OLS imposes symmetry assumptions that may underrepresent tail risks periods of extreme congestion or exceptionally high performance. Future research should address these limitations by extending the model across multiple terminals and by integrating moderating and mediating factors such as buffer capacity, yard allocation strategies, berth planning, or weather disruptions. Methodologically, more advanced approaches offer promising avenues: • Quantile regression, to capture how delays affect low-productivity versus highproductivity situations differently. • Threshold or regime-switching models, to detect operational breakpoints (e.g., when delays surpass critical values). • Reinforcement Learning and LSTM-based predictive control, to transition from reactive delay measurement toward proactive orchestration of cranes, vehicles, and yard assets. Strategically, the findings reaffirm that reducing delays remains one of the most tractable and impactful pathways for enabling smarter, cleaner, and more resilient port operations. ISSN: 2658-8455 Volume 6, Issue 13 (2025), pp. 324-341. © Authors: CC BY-NC-ND . 340 www.ijafame.org References: (1). Alamoush, S., Ballini, F., & Ölçer, A. (2020). Deep learning tools to improve maritime transportation sustainability: A review. Sustainability, 12(2), 823. https://doi.org/10.3390/su12030823 (2). Al-Fatlawi, M. A., & Motlak, M. (2023). IoT-based digital transformation in port operations: A framework for smart maritime logistics. Journal of Marine Science and Engineering, 11(3), 612. https://doi.org/10.3390/jmse11030612 (3). Brunila, O.-P., Kunnaala-Hyrkki, V., & Inkinen, T. (2023). Sustainable small ports: Performance assessment tool for management, responsibility, impact, and selfmonitoring. Journal of Shipping and Trade, 8(1), 14. https://doi.org/10.1186/s41072023-00142-z (4). Brooks, M. R., & Cullinane, K. (Eds.). (2006). Devolution, port governance and port performance. Elsevier. (5). Dalaklis, D., Kotios, A., & Ntalos, A. (2023). Digital transformation in maritime transport and ports: Trends, technologies and challenges. Journal of Shipping and Trade, 8(1), 4. https://doi.org/10.1186/s41072-023-00132-1 (6). Dewita, W. A., Yen, B. T. H., & Burke, P. F. (2018). Evaluating seaport efficiency using a two-stage DEA approach: Evidence from Indonesia. Asian Journal of Shipping and Logistics, 34(2), 113–122. https://doi.org/10.1016/j.ajsl.2018.06.002 (7). El Imrani, A., El Ghmari, A., & El Mghari, M. (2021). Efficiency assessment of Moroccan container terminals using DEA. International Journal of Maritime Engineering, 163(A4), 45–56. (8). Gattuso, D., Pellicanò, D. S., & Cassone, G. C. (2023). Policies and models for efficient and eco-sustainable ports. The Open Transportation Journal, 17, e26671212265181. https://doi.org/10.2174/0126671212265181230921103400 (9). Hardianto, A., Pradana, M., & Santosa, B. (2023). Key performance indicators for sustainable port operations: A systematic literature review. Sustainability, 15(6), 5520. https://doi.org/10.3390/su15065520 (10). Haralambides, H. E. (2017). Structure and operations in maritime transport: An overview of port performance determinants. Maritime Economics & Logistics, 19(3), 455–478. https://doi.org/10.1057/mel.2017.5 (11). Hu, Z., Diabat, A., & Zhao, X. (2013). Optimal buffer allocation for quay cranes in container terminals. Transportation Research Part E: Logistics and Transportation Review, 51, 17–34. https://doi.org/10.1016/j.tre.2012.12.002 (12). Hsu, W.-K. K., Luo, M., & Zhang, H. (2023). Big-data-driven port performance analysis: Methods and applications. Maritime Policy & Management, 50(1), 1–21. https://doi.org/10.1080/03088839.2022.2074671 (13). Kishore, L., Pai, Y. P., Ghosh, B. K., & Pakkan, S. (2024). Maritime shipping ports performance: A systematic literature review. Discover Sustainability, 5, 108. https://doi.org/10.1007/s43621-024-00299-y (14). Lee, E. S., & Song, D. W. (2023). Post-pandemic port operations: Digitalization, automation, and resilience challenges. Maritime Economics & Logistics, 25(2), 247– 266. https://doi.org/10.1057/s41278-022-00231-4 (15). Li, J., Wang, S., & Xu, Y. (2022). Quantile regression for maritime logistics: Modeling productivity under delay uncertainty. Transportation Research Part E, 162, 102706. https://doi.org/10.1016/j.tre.2022.102706 (16). Luo, M., & Wu, J. (2020). Vessel scheduling and quay crane operations: Minimizing makespan in container terminals. Transportation Research Part B, 132, 63–79. https://doi.org/10.1016/j.trb.2019.12.013 Zakaria ELKHARMALI & Ouail EL KHARRAZ. Investigating Maritime Ports’ Operational Performance Determinants: An Empirical Study of Operational Delays and Moroccan Container Terminal Productivity 341 www.ijafame.org (17). Molavi, A., Lim, S., & Race, B. (2020). A framework for evaluating port performance in the era of automation and digitalization. Maritime Policy & Management, 47(2), 139– 155. https://doi.org/10.1080/03088839.2019.1697871 (18). Mishrif, A., & Khan, S. (2023). Pandemic-driven disruption and digitalization in seaports: Lessons from COVID-19. Research in Transportation Business & Management, 46, 100907. https://doi.org/10.1016/j.rtbm.2023.100907 (19). Naeem, D., Gheith, M., & Eltawil, A. (2023). A comprehensive review and directions for future research on the integrated scheduling of quay cranes and automated guided vehicles and yard cranes in automated container terminals. Computers & Industrial Engineering, 179, 109149. https://doi.org/10.1016/j.cie.2023.109149 (20). Pallis, A. A., Vitsounis, T. K., & de Langen, P. W. (2010). Port economics, policy and management: Review of research. Maritime Economics & Logistics, 12(1), 1–41. https://doi.org/10.1057/mel.2009.22 (21). Paraskevas, A., Madas, M., Zeimpekis, V., & Fouskas, K. (2024). Smart ports in Industry 4.0: A systematic literature review. Logistics, 8(28). https://doi.org/10.3390/logistics8010028 (22). Paternina-Arboleda, C. D., González-Ramírez, R., & Barrera, B. N. (2023). Real-time optimization framework for smart port operations using digital twins. Logistics, 7(4), 87. https://doi.org/10.3390/logistics7040087 (23). Peng, W., Bai, X., Yang, D., Yuen, K. F., & Wu, J. (2022). A deep learning approach for port congestion estimation and prediction. Maritime Policy & Management. https://doi.org/10.1080/03088839.2022.2057608 (24). Roy, D., Goswami, A., & Bandyopadhyay, S. (2020). Performance evaluation of container terminals under varying buffer conditions. Maritime Transport Research, 2, 100009. https://doi.org/10.1016/j.martra.2020.100009 (25). Singh, S., Emrouznejad, A., & Pratap, S. (2025). Achieving net zero: Enhancing maritime port efficiency through multi-objective integrated DEA-MORCOS. Case Studies on Transport Policy, 22, 101597. https://doi.org/10.1016/j.cstp.2025.101597 (26). Song, Z.-Y., Lin, C.-W., Feng, X., & Lee, P. T.-W. (2024). An empirical study of the performance of the sixth-generation ports model with smart ports with reference to major container ports in mainland China. Transportation Research Part E: Logistics and Transportation Review, 184, 103460. https://doi.org/10.1016/j.tre.2023.103460 (27). UNCTAD. (2021). Review of Maritime Transport 2021. United Nations Conference on Trade and Development. (28). Xinyan, S., Qiang, M., & Lijun, Z. (2014). Impact of buffer capacity on quay crane scheduling performance. Procedia Engineering, 79, 216–225. https://doi.org/10.1016/j.proeng.2014.06.343 (29). Yen, B. T. H., Huang, M.-J., Lai, H.-J., Cho, H.-H., & Huang, Y.-L. (2023). How smart port design influences port efficiency: A DEA-Tobit approach. Research in Transportation Business & Management, 46, 100862. https://doi.org/10.1016/j.rtbm.2023.100862 (30). Yue, X., Chen, Y., & Li, Z. (2021). Integrated scheduling of quay cranes and automated vehicles in container terminals. Computers & Industrial Engineering, 159, 107511. https://doi.org/10.1016/j.cie.2021.107511 (31). Zhen, L. (2015). Container terminal operations management: Review and future directions. Maritime Economics & Logistics, 17(1), 1–25. https://doi.org/10.1057/mel.2014.23