KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/kmitl/1

Browse

Settings

Author: search.filters.author.Dahlan, Ahmad Faisal

Search Results

Now showing 1 - 3 of 3
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Crowdsourced adaptive vehicle routing framework for Last-Mile delivery in dynamic traffic environments
    (2025-01-01)
    Dahlan, Ahmad Faisal
    ;
    Cheng, Chen Yang
    ;
    Sae-chai, Pornkanok
    ;
    Jientrakul, Ranon
    ;
    Yuangyai, Chumpol
    Last-Mile Delivery (LMD) operations are significantly impacted by real-time traffic disruptions, leading to delays and increased costs. While traditional vehicle routing problem (VRP) model struggle to adapt to dynamic traffic environments, crowdsourced data from social media platforms presents a valuable source of real-time traffic data. This research proposes Crowdsourced Adaptive Vehicle Routing Framework (CAVRF) that integrates crowdsourced social media data into the VRP model for enhanced efficiency. The framework employs a machine learning model to classify tweets based on impact severity and effectively filtering relevant traffic information. Furthermore, a mathematical model known as the Adaptive Traffic VRP (AT-VRP) has been developed to accommodate the integration of social media data with the VRP model. The framework’s effectiveness is demonstrated through a case study using a package delivery network in Jakarta with various levels of traffic disruptions. The findings suggest that integrating crowdsourced social media data into AT-VRP significantly improves efficiency by avoiding any road closure. CAVRF offers a cost-effective and efficient solution to the dynamic challenges inherent in LMD.
      3
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    T5-Based Named Entity Recognition for Social Media: A Case Study for Location Extraction
    (2024-01-01)
    Dahlan, Ahmad Faisal
    ;
    Yuangyai, Chumpol
    Social media platforms have emerged as invaluable sources of real-time information, particularly during emergencies and events. Efficient and accurate extraction of location names from this unstructured text data can significantly enhance response efforts. This study investigates the performance of two models, T5 and SpaCy, for extracting location names from 5554 Indonesian-language tweets related to traffic conditions. The T5 model, leveraging its Transformer architecture and extensive pre-training, achieved a significantly higher accuracy of 95% in training and 93% in testing compared to SpaCy's 45% and 41% respectively. This disparity highlights T5's superior ability to handle complex language patterns and indirect location references often found in social media text. Conversely, SpaCy's reliance on Convolutional Neural Networks (CNNs) poses limitations in effectively processing diverse location representations and non-local text patterns. The results demonstrate the potential of T5 as a powerful tool for location extraction in social media analysis, with significant implications for improving disaster response and public safety efforts.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    A Last-mile Facility Allocation Problem with Traffic Congestion and Natural Disaster
    (2023-01-01)
    Dahlan, Ahmad Faisal
    ;
    Yuangyai, Chumpol
    Managing distribution in urban areas can be much more challenging in the twenty-first century. The population density carries accessibility issues to education, healthcare, infrastructure, and services through traffic congestion and pollution. It is worth noting that some areas simultaneously deal with traffic congestion and natural disasters. This issue may interfere with last-mile logistics, mainly when extreme conditions occur. One way to overcome this issue is by placing distribution facilities in unaffected areas to maximise last-mile logistics despite extreme conditions. However, the facility location problem regarding traffic congestion and natural disasters remains limited. This paper proposes Expected coverage problems with traffic congestion and natural disaster (ECP-CN), the extension model of facility location problems involving traffic congestion and natural disasters. ECP-CN consists of flexibility where the facilities will be in non-disaster areas rather than congestion-free areas whenever the ideal condition is not met. The algorithm used to solve the problem is a Genetic Algorithm (GA). The algorithm has converged to an optimal solution, showing a high initial demand coverage, subsequent improvements, and significant fluctuations. Further exploration of parameter settings or diversity-enhancing mechanisms may be optional. This result indicated that GA could locate the facilities in non-disaster areas while maintaining maximum demand coverage in all sample sizes.