This study is devoted to the use of varied analytical methods to elucidate the complex relationship between normalized difference vegetation index (NDVI) and wheat production in Taounate, Morocco based on MODIS Satellite data. Linear regression (LR), with a coefficient of determination (R²) of 0.93, provided a solid basis, while the decision tree (DT) showed significant performance with an R² of 0.81. Support vector regression (SVR) performed well with an R² of 0.96, highlighting its ability to capture the non-linear nuances of the data. Given the complexity inherent in the observed relationships, characterized by non-linear variations, we opted for a combined approach. K-means, closely linked to SVR, was integrated for its ability to identify homogeneous subgroups in the data (R2 up to 0.98). This combination made it possible to circumvent the limits of strictly linear methods, thus reinforcing the robustness of our analysis. These results underline the capacity of the chosen methodology to decode the interactions between NDVI and wheat production in the complex context of Taounate. By providing clear and nuanced perspectives, this study helps inform agricultural decisions and build resilience to climate challenges in the region.

Mass customization emerged as a strategic response to the increased consumer interest in personalized products, combining the efficiency of mass production with the flexibility of individual customization. Yet, this manufacturing paradigm introduced significant complexities in demand forecasting, where traditional methods struggled to account for the high variability in customer preferences. The lack of accessible, high-quality datasets that capture real customization behavior makes this challenge more difficult. To address this gap, this study develops and validates a large-scale synthetic dataset simulating 500,000 laptop sales. Each sale corresponded to a customized product configuration tailored to an individual customer profile, thereby reflecting the dynamics of real-world mass customization contexts. The dataset was generated using a structured rule-based framework, supported by artificial intelligence (AI)-assisted reasoning and grounded in domain knowledge and manufacturing literature. This framework was designed to reproduce realistic customer behavior, configuration patterns, and seasonal demand fluctuations. As a result, the generated data captured heterogeneous customer profiles and varied product configurations. Its realism and coherence were assessed through statistical distribution analysis, logical consistency checks, and seasonal pattern verification. The validation results provide a reliable basis for evaluating and benchmarking demand forecasting models. Overall, this work shows that combining rule-based design with AI-assisted reasoning is effective for the generation of realistic synthetic datasets for demand forecasting in mass customization contexts where real data is unavailable. Received: 16 January 2026 | Revised: 22 April 2026 | Accepted: 20 May 2026 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The data that support the findings of this study are openly available in Kaggle at https://doi.org/10.34740/kaggle/dsv/15258985. Author Contribution Statement Nouhaila El Assad: Conceptualization, Methodology, Software, Formal analysis, Data curation, Writing – original draft, Visualization. Kawtar El Haouti: Methodology, Validation. Soumaya Zayrit: Software, Data curation. Salah-eddine Mokhlis: Validation, Formal analysis. Mohamed Rhouzali: Formal analysis, Investigation. Najat Messaoudi: Validation, Writing – review & editing, Supervision, Project administration.