Practical Examples of How AI-Powered Short-Term Forecasting Drives Fleet Efficiency

The daily challenges faced by road freight operations departments are numerous, multifaceted and fast-changing. In an environment characterised by the unexpected, the reliability of forecasts can be make-or-break for overall fleet efficiency.

Short-term resource management is a critical but often neglected component of fleet operations. Operations managers have to oversee the dynamic allocation and reallocation of fleet resources to meet immediate demands, make agile and effective decisions to respond to last-minute changes and manage any disruptions in the logistics chain.

Integrating machine learning (ML) and artificial intelligence (AI) into short-term demand forecasts can significantly enhance a logistics service provider’s (LSP) ability to adapt to dynamic conditions. These technological advancements boost fleet productivity and efficiency via various practical applications. 

Here, we’ll discuss some real-life use cases of ML and AI-powered short-term demand forecasting, demonstrating their tangible impact on fleet efficiency. 

The four key factors impacting fleet efficiency in the road freight space

Fleet efficiency isn’t a two-dimensional concept but a combination of several critical components that collectively enhance the performance and cost-effectiveness of transport operations. Before jumping into the examples, we’ll quickly outline the specific metrics that drive overall fleet efficiency:

• Resource management: Effective resource management maximises volume demand with minimal resource waste, uses the exact number of vehicles needed to meet demand, and reduces unnecessary costs and environmental impact.
• Optimisation: Optimal resource utilisation across internal and outsourced fleets involves maximising vehicle capacity, minimising empty or underutilised trips, and aligning operations with sustainability objectives to reduce both costs and carbon emissions.
• Flexibility: Flexible fleet management allows for the dynamic adjustment of fleet capacity to actual conditions. This enables operations to unexpected changes in demand and gain a competitive advantage in a market heavily impacted by seasonality.
• Agility: In a dynamic and changeable environment like road freight transport, operations teams should utilise advanced dashboards and automated tools to monitor and adjust operations continuously. This reduces manual intervention and enhances efficiency, enabling swift adaptation to changing circumstances without disrupting service quality.

Practical applications of AI tools

The image below shows insights from Ontruck’s AI/ML tool, followed by practical examples of what can be done with such insights. This demonstrates how AI and ML solutions can impact the four quantifiable factors mentioned above, generating accurate short-term forecasts and driving fleet efficiency.

1. Enhance resource management and optimise demand forecasting

Real-time monitoring systems that integrate forecasts enable continuous visibility of forecast, demand and capacity,  allowing operations teams to respond to deviations between predicted and actual demand immediately. This capability ensures that service levels are consistently met and operational efficiency is maintained. 

Constant access to a real-time forecast is transformative, especially when paired with demand and capacity data. This continuous stream of information ensures ops managers can maintain a clear and current view of their fleet, allowing for more informed and strategic decision-making throughout the day. 

Hands-on operational example

For example, in a traditional working model, shipment requests are received in the morning. In the afternoon, the ops team will plan routes and assign shipments to vehicles. It’s at that moment that the team might notice there are too many or too few vehicles. This leads to one of two scenarios:

• There’s a lack of capacity, resulting in a frantic search for vehicles. Either the team finds extra capacity at a last-minute premium, or they don’t. The latter implies the customer has been failed and the business lost.
• There is excess capacity and contracted vehicles have to be paid even if they go empty.

Traditionally, the road freight industry is very reactive: solutions are sought when it's too late. ML-enabled forecast data provides insights into deviations from previous days’ shipment volumes by generating forecasts and assessing capacity. If forecasted volumes exceed capacity, external fleet assistance can be sought proactively. Conversely, if forecasted volumes are below capacity, external carriers can be informed of reduced need, thus saving costs. 

Forecast data allows ops teams to see where they stand and which customers are increasing their shipment demand. Armed with this intelligence, the team can formulate a proactive action plan. This approach enhances customer service levels while optimising operational expenses. 

2. Enhance agility and flexibility with real-time data visibility 

Deploying ML and AI tools for real-time fleet management allows for the automation of complex decision-making processes. These tools can analyse vast amounts of data, enabling logistics operators to make informed decisions quickly, thus improving daily operational management and supporting agile responses to unexpected changes.

Hands-on operational example

Algorithms can analyse large amounts of data regarding shipments, locations, and cargo characteristics. With this information, they can forecast the types of vehicles needed each day, tailored to specific cities, route types and various freight operations (like full truckload, less than truckload, international, domestic, and last mile).

This forecasting helps identify changes in vehicle requirements and assesses the number of assigned shipments and potential shortages. Simultaneously, it demonstrates how the market is evolving and the net revenue achieved. It enables real-time monitoring of profit margins based on vehicles and routes, facilitating timely actions to maximise profitability.

3. Drive fleet efficiency without re-inventing the wheel

Integrating AI and ML forecast results with existing management systems can greatly increase fleet efficiency. This integration facilitates more seamless operations, reduces the need for manual interventions, and enhances overall operational responsiveness.

Hands-on operational example

For example, we can automatically input the anticipated number of vehicles into a given system. Specifically, our forecast module can seamlessly integrate with our routing tool, enabling the pre-booking functionality. Within our operations platform, the pre-booking tab presents recommendations for reserving slots, guided by the forecast and driver availability.

To provide a straightforward illustration, picture a forecast indicating a requirement for 10 trailers; consequently, the system triggers 10 pre-bookings automatically. The team can then assess and accept these recommendations, guaranteeing ample capacity to manage demand efficiently. This functionality empowers fleet managers to swiftly respond to forecast analysis, thereby ensuring streamlined operations.

Data-driven solutions informed by hands-on experience

The strategic implementation of ML and AI forecasting methods represents a transformative leap forward in fleet management. AI and ML forecasting can significantly improve fleet efficiency by enabling more accurate and timely decision-making in both short-term resource management, as we’ve seen here, and medium-to-long-term capacity planning.

In this article, we've explored practical approaches to achieving greater fleet efficiency based on firsthand experience. It's important to highlight that our solutions have been developed through real-world challenges encountered within our freight business division, Ontruck Transport Services.

Strategies and techniques that harness the power of AI and ML in short-term fleet management not only streamline operations but also enable you to navigate unexpected challenges adeptly. Schedule your complimentary demo today and discover how our cutting-edge solutions can optimise your fleet management process.

‍