Shipping Storage Risk Prediction using Machine Learning Model

case study - gen ai - logistics and shipping
Executive Summary:

In the shipping and logistics industry, demurrages (port fees for delayed shipments) stand as both crucial revenue streams and potential financial burdens for businesses. The accurate prediction of demurrage costs is a multifaceted challenge, influenced by dynamic factors such as customer behavior, terminal operations, port congestion, and weather conditions. Failure to effectively anticipate demurrages not only leads to financial losses but also undermines operational efficiency and customer satisfaction.

Recognizing these complexities, our team at Factspan embarked on a comprehensive development journey to address the challenges posed by demurrage prediction. Leveraging our expertise in AI and machine learning, we meticulously designed and implemented an innovative demurrage risk assessment model. Our approach aimed not only to forecast demurrage costs accurately, but also to provide actionable insights for proactive decision- making.

About the Client

The client is a leading player in the shipping and logistics sector, offering freight transportation services across various industries. Seeking to optimize operations and enhance customer satisfaction, the client required a solution to predict and manage demurrage costs effectively.

Business Challenge

The organization faced several challenges in predicting demurrage costs. Reactive responses due to the absence of predictive models led to increased operational costs and inaccurate delay assessments. Data silos and limited data sharing hindered holistic analysis, impacting competitiveness and cost- effectiveness. Inefficiencies and missed optimization opportunities further compounded the challenges.

Our Solution

To address the intricate challenges of demurrage prediction, the Factspan team adopted a multi-faceted approach:

Data Extraction and Preparation: Factspan meticulously gathered raw data from diverse sources including Maestro, NSCP, MODS, and APIs. This data was then meticulously cleaned, normalized, and integrated to ensure its suitability for analysis.

Advanced Data Analysis and Predictive Modeling: Factspan utilized advanced statistical techniques to analyze historical data from January 2021 to December 2023. By leveraging Catboost-based ML algorithms, predictive models accurately forecast demurrage risks and chargeable days.

Model Deployment and Integration: The Factspan team seamlessly deployed models with ML Flow, integrating predictions into workflows for stakeholder access. Foreline agents received insights via SharePoint, while customer experience agents accessed a user-friendly interface for real-time support.

Tools used

Azure Databricks, Azure Data Lake, Azure Workflow, ML Flow, Power BI, Confluence, Jira

Business Impact

The implementation of our demurrage risk assessment system resulted in significant business outcomes:

  • Cost Reduction: Achieved a 20% reduction in overall demurrage expenses
  • Operational Efficiency: Realized a 15% increase in efficiency and productivity
  • Revenue Optimization: Attained a 10% increase in revenue optimization
  • Improved Satisfaction: 25% increase in customer satisfaction and loyalty
Most Popular
Gen ai usecases

Gen AI-Infused Business Analytics for En...

Empowering data team

Boosting Efficiency with Gen AI through ...

Enhancing Developer Output with Gen AI f...

Streamlining Sales Cover Image

Streamlining Sales Queries with AI Bots ...

Download Case Study

    Work Email*

    Company Name (Optional)

    Scroll to Top