At the same time, the evidence should be interpreted considering the sample size and regional scope, and would benefit from external validation on larger, multi-region datasets curated under common taxonomies. Future work should explore region-specific variations in accident patterns to improve model applicability in diverse construction environments. Overall, SHAP analysis strengthens the interpretability of the model, providing construction managers and safety officers with clear guidance for implementing effective safety measures.
- Plus, if your virtual design and construction team is stretched thin, this is the kind of tool that scales site capture without scaling headcount.
- This study provides actionable insights into improving safety management practices in construction, particularly in safety-critical environments.
- For this study, the value of k (number of neighbours) was determined empirically by testing a small range of candidate values (3–9).
- The area under the ROC curve (AUC-ROC) is a critical metric that quantifies the model’s ability to correctly classify incidents at different severity levels, with values closer to 1.0 indicating superior performance.
- Analysing MIMIC III data, a random forest (RF) model achieved 90% accuracy in predicting LOS, demonstrating the potential of ML for optimizing resource allocation and health safety policies in construction.
Collectively, the feature importance patterns explain the marked improvement in XGB performance when NOI is incorporated, as it captures critical distinctions among incident types that strongly determine severity. Boosting algorithms, for example, iteratively refine the model by focusing on hard-to-classify instances, ensuring improved performance on imbalanced and noisy datasets. To provide a more thorough evaluation, Micro-Average AUC scores were used to compare models, as they aggregate performance across all severity levels. Table 7, which focuses exclusively on SOI Level 5 (fatal incidents), provides deeper insights into the models’ ability to https://drpostdoc.com/what-do-you-need-to-work/ predict the most critical severity class. To ensure a more balanced evaluation, precision, recall, and F1-score were also calculated to provide deeper insight into model performance for SOI (including NOI). In this study, SHAP was used to quantify the relative influence of each input on predicted outcomes and to summarize feature contributions across the dataset, supporting transparent interpretation of the learned models.
To further understand the model performance, confusion matrices were generated for each ML model to visualize how accurately they classified different severity levels. In this study, multiclass ROC curves were generated using the One-vs-Rest (OvR) approach, where the classification performance for each severity level was assessed separately (Fig. 5(a-f)). DT performed moderately well, maintaining a balanced precision-recall trade-off but was less effective than RF or boosting models.
Models like RF and XGB aggregate multiple DTs, each specializing in different regions of the feature space, allowing them to handle complex decision boundaries effectively. The dataset contains non-linear relationships and interactions between explanatory variables (e.g., the combined effect of PPE usage, safety training, and incident location), which are better captured by ensemble methods. These models showed strong predictive capability for high-severity https://housebru.com/website-development-and-promotion-for-construction-companies-and-developers.html incidents (Severity Level 5), minimizing errors in distinguishing critical safety events.
Transforming how projects are built and managed to increase efficiency and profits
A machine learning model could learn from historical patterns in thousands to millions of similar jobs to assess patterns and flag poor outcomes, such as margin drawdowns, project delays, project rework. Wrike’s Jamie Eckmier dives into the complexities of product lifecycle management, the promise of AI, and practical tips for manufacturers. Build an effective construction management plan that helps you control budgets, timelines, and communication protocols. In our exhaustive guide to construction project management, you’ll find construction management basics, tips and tricks to ensure consistent success, and the tools you’ll need along the way. The real risks are mundane things, such as leaking sensitive project data into tools you don’t control and blurring responsibility when a bad suggestion quietly influences a cost, scope, or safety decision.
Machine learning framework for holistic evaluation of construction projects using the project performance index
- Models were trained on the SOI and NOI dataset using an 80/20 train-test split with a fixed random state of 123.
- These findings align with the overall evaluation in Table 5, further validating the reliability of boosting-based models like XGB and GB for accurately identifying high-risk events in construction safety (Table 8).
- Given the results, XGB is recommended as the most suitable model for implementing proactive safety assessment systems, given its superior performance in distinguishing incident severity and handling complex datasets.
- While RF benefited from its ensemble learning capability, it still struggled in cases where severity levels overlapped.
Models were trained on the SOI and NOI dataset using an 80/20 train-test split with a fixed random state of 123. In the realm of injury severity analysis, DTs offer a robust approach to exploring the complex interplay between various factors and the resulting injury outcome. The primary objective of a linear SVM is to determine the most suitable separation boundary that optimizes the margin between two distinct classes.
Pick one pain point with clear ROI, like RFI drafting and document search or progress capture, and pilot it with one project team before scaling. Wrike provides construction managers with a single, grounded platform to manage a job, where tasks, conversations, and documentation remain linked, and the system surfaces what needs your attention, rather than requiring you to dig for it. If you want AI to do real work for you, start by making the workflow predictable with Wrike AI. What do you lean on when the day-to-day grind of running construction projects starts swallowing the bigger business decisions? ClaimMaster.ai positions itself as construction-specific AI for claims, including assistants for delay, quantum, and legal work, and emphasizes compliance and data handling in its materials.
The dataset exhibited a significant class imbalance, with fatalities (SOI Level 5) accounting for only 11% of the total incidents. The distribution (Fig. 1) highlights that most incidents fall into moderate to severe categories (SOI Levels 2–4), while fatal incidents (SOI Level 5) represent a smaller yet critical portion of the dataset (11%). The table summarizes the mean, standard deviation, minimum, and maximum values, which help readers understand the distribution and scale of the encoded variables used in model training. To provide a clear understanding of the modelling dataset after encoding (Sect. “Encoding categorical data”), Table 3 presents the descriptive statistics of the 14 explanatory variables and the target (SOI). These exclusions were necessary to ensure data quality and reliability, leaving 203 incidents as the final dataset used for model development and analysis. Section “Results and discussion” presents the results and discussion, covering model validation, performance evaluation, feature importance analysis, and SHAP-based interpretability.
Plus, if your virtual design and construction team is stretched thin, this is the kind of tool that scales site capture without scaling headcount. Rather than relying on one baseline, AI tools can generate alternative sequences, test constraints, and reveal which options hold up best once work begins. Civils.ai pitches itself as AI for PDF/CAD takeoffs, estimation, and quantity surveying, with additional positioning around extracting data from drawings like schedules, notes, tables, and specifications. On most jobs, the real bottleneck isn’t steel or concrete — it’s the paperwork stuck in someone’s inbox. AI features like Construction IQ, Autodesk Assistant, photo auto-tagging, and predictive risk scoring help teams sift through drawings, photos, and project data to surface issues earlier and reduce rework.
However, the regularization parameter (C) and kernel parameter (gamma) were kept at default values https://pagemakers.net/category/environment-and-sustainability/ due to the study’s focus on comparative evaluation rather than exhaustive hyperparameter tuning. For this study, the RBF kernel (Radial Basis Function) was used, as it is well-suited for datasets with non-linear relationships. XGB and RF were chosen for their ability to handle non-linear relationships and feature interactions, which are critical in analysing the complex factors contributing to safety incidents. Such imbalance can bias ML models toward majority classes, reducing their ability to predict minority outcomes like fatalities.
The modelling is complemented by SHAP-based interpretability to identify the most influential factors for decision-making. This study analyses 203 construction safety incidents from Saudi Arabia (primarily Makkah and Riyadh) to develop predictive models for both the NOI and the SOI using 14 explanatory variables. While this study focused on the Saudi Arabian construction industry, its methodology can be adapted for other regions by incorporating local climate, workforce demographics, and regulatory frameworks. The inclusion of NOI as a feature not only enhances interpretability but also explains the substantial increase in XGB’s SOI prediction accuracy (89%) when it is incorporated. Additionally, the “Date of Incident” revealed temporal patterns, with certain months exhibiting higher risks, providing opportunities for seasonal safety planning. These findings underscore the need for weather-adaptive safety planning, including schedule adjustments, drainage improvements, and workforce training for hazardous weather.