Fine-Tuning Dynamic Route Optimisation to Redefine Efficiency in Logistics

The European logistics landscape is facing a mounting challenge: the relentless rise of general haulage costs. Recent reports from the Germany-based Federal Association of Forwarding and Logistics paint a stark picture, revealing a 3.2% increase in shipment-related processing costs in just the latter half of 2023. This surge, driven by escalating personnel, material, and operational expenses, is putting immense pressure on the industry's bottom line. In context, dynamic route optimisation emerges as a key tool to alleviate the pressure. 

Dynamic route optimisation leverages real-time data on delivery time windows and even vehicle capacity and customer preferences. It then applies intelligent algorithms to continuously calculate and adjust the most efficient routes. By dynamically adapting to changing circumstances, it goes beyond traditional static route planning, offering the potential to significantly trim transportation costs, minimise delays and enhance overall operational efficiency. 

Several different techniques fall under the umbrella of dynamic route optimisation, and in the following sections, I’ll outline the features of some of the most effective methods in more detail.

The difference between static and dynamic route optimisation

In the transportation industry, routing is the process of determining the most efficient way to use vehicles and drivers to complete a set of deliveries or pickups. Traditionally, routing has been a static process. This means that all orders, vehicles and constraints need to be known well in advance. The analysis is typically done once, leaving little flexibility for changes or unexpected events. 

These constraints can include factors like the number of vehicles in a fleet (whether it's fixed or variable), staff availability, specific delivery time windows (e.g., all pickups must happen in the morning) and customer preferences. The optimisation goal can also differ depending on the company's priorities. It could be minimising costs by finding the cheapest routes, reducing the distance travelled or carbon emissions, or improving customer service by minimising delays and ensuring on-time deliveries.

However, these factors are not static quantities but constantly shifting variables. Dynamic route optimisation offers a more adaptable approach that reflects these conditions. It continuously updates and adjusts routes as new orders arrive, traffic conditions change, or other unforeseen circumstances arise. By incorporating real-time data on vehicle locations, warehouse availability, and more, dynamic routing algorithms provide a level of flexibility and responsiveness that was previously impossible.

Advanced routing techniques

Constantly shifting variables can throw even the most meticulously planned routes into disarray. To navigate this ever-changing landscape, logistics companies are turning to dynamic route optimisation, a suite of algorithms and technologies designed to adapt and evolve in real-time.

At the core of these technologies lies a Python-based API enriched with machine learning algorithms trained on vast datasets from real-world transport operations. This allows for the creation of highly accurate and predictive models that can anticipate and respond to changing conditions on the fly. 

These varied approaches range from sophisticated algorithms that incorporate real-time traffic data and points of interest (POI) to hybrid models that combine fixed and dynamic routing for maximum fleet efficiency. Some focus on seamlessly adding new stops to existing routes, while others continuously monitor and adjust routes based on live updates from the field. Let’s examine each of these dynamic route optimisation methods in greater detail.

1. Advanced modelling incorporating POI functionality

Points of Interest (POI) are systems that gather detailed metadata about various locations, enhancing the accuracy and efficiency of delivery route planning. This metadata encompasses a wide range of factors that influence the execution of deliveries, including warehouse operating hours, specific handling requirements, vehicle restrictions, and more. 

By incorporating this granular level of information, POI-enabled dynamic route optimisation can generate optimised routes with exceptional precision. Key factors that can be integrated into POI data include:

• Obligatory and mandatory break times for drivers
• Warehouse opening hours and loading/unloading protocols
• Cost variations based on vehicle type
• Highly customisable cost functions
• Custom handling times per warehouse

In many logistics service providers (LSPs) and transport companies, fleet and operations teams often rely heavily on the knowledge of experienced personnel who understand the nuances of various delivery locations. However, this reliance can lead to disruptions when these individuals are unavailable. POI integration effectively translates the expertise of seasoned fleet managers into actionable data.

For instance, if a specific warehouse closes at 6 pm, the POI-enabled routing algorithm will ensure that deliveries to that location are scheduled before the closing time, preventing wasted trips and potential delays. By automating this type of decision-making, POI systems streamline operations, reduce the risk of human error, and ultimately enhance customer service.

2. Last-mile hybrid routing

Last-mile delivery often poses significant logistical challenges, particularly in urban environments. Hybrid routing addresses this by intelligently combining two modes of transportation: vans for longer distances between clusters of deliveries and foot travel for shorter distances within each cluster.

This approach involves two distinct phases:

1. Cluster formation: The initial step is to divide the delivery area into clusters. The goal is to optimise the overall delivery time by considering factors like travel time between deliveries via both van and foot, as well as the time required to park the van at each stop. This ensures that the combination of driving and walking results in the most efficient delivery process.
2. Transport between or within clusters: Once the clusters are defined, the routing algorithm solves the problem as usual, but with the clusters acting as waypoints. Travel times between deliveries are adjusted based on whether they involve driving or walking within a cluster.

The primary advantage of hybrid routing is a reduction in overall delivery time. However, the optimal cluster configuration is highly context-dependent. Driving between most stops might be more efficient in areas with low traffic congestion and ample parking. Conversely, walking between deliveries within a well-chosen cluster in dense urban areas with heavy traffic and limited parking could save significant time. By adapting to the specific characteristics of each delivery area, hybrid routing maximises efficiency and ensures timely deliveries in diverse urban settings.

3. Hot routing

Hot routing addresses a common scenario in logistics: drivers with partially filled schedules due to individual spot orders. When a new order arrives that aligns well with an existing assignment, the system automatically identifies this opportunity and proposes a route upgrade to the driver. This new, combined route is optimised for efficiency, ensuring that the driver's working time is utilised effectively.

At its core, hot routing is a continuous optimisation process. It constantly seeks the best possible solution for all orders, whether they are already assigned to a driver or not. The system continuously evaluates the existing schedule against incoming orders, identifying opportunities to consolidate routes and improve efficiency.

The hot routing process unfolds as follows:

1. Upgrade offer: When a new order arises that complements an existing route, the system generates an upgrade offer for the assigned driver.
2. Route optimisation: The new, combined route is automatically optimised, taking into account all relevant constraints and factors, such as distance, time windows, and vehicle capacity.
3. Cost recalculation: The system recalculates the cost of the upgraded route, adjusting the driver's earnings and the overall profit margin for the combined delivery.
4. Visual verification: A map visualisation of the new route is provided, allowing fleet managers to easily review and verify the proposed changes.

Importantly, fleet managers retain full control and can manually modify routes if necessary, ensuring flexibility and responsiveness to unforeseen circumstances.

By dynamically adapting to new orders and maximising driver utilisation, hot routing significantly improves operational efficiency, reduces costs and enhances driver satisfaction by providing them with optimised routes and increased earning potential.

4. Always-on mode

In always-on mode, the routing algorithm operates continuously, constantly analysing and refining routes in response to real-time data. Whether a new order arrives, traffic conditions change, or unforeseen disruptions occur, the system dynamically recalculates and proposes improved routes to maintain peak operational efficiency.

Similar to hot routing, the always-on mode considers the entire pool of orders, including those already assigned to drivers and those pending assignments. This holistic perspective allows for superior optimisation, as the algorithm can identify opportunities to consolidate routes, minimise empty miles and adapt to changing circumstances on the fly.

Always-on dynamic route optimisation ensures effective resource utilisation, timely deliveries, and high customer satisfaction by continuously adapting to new information and optimising across all orders. This approach not only enhances operational efficiency but also provides a competitive advantage in an industry where responsiveness and agility are paramount.

Combine solutions to create a bespoke solution

Ontruck customer data shows that advanced routing techniques reduce the overall distance required to fulfil orders by an impressive 23%. This translates directly into cost savings or, alternatively, frees up capacity for increased revenue generation. Additionally, the increased precision in collection and delivery processes facilitated by dynamic route optimisation leads to improved service levels and a staggering 2.2 times increase in productivity.

Dynamic route optimisation is not just a technological advancement; it's a game-changer for the logistics industry. By harnessing real-time data, intelligent algorithms and advanced features like POI integration and hybrid routing, companies can achieve remarkable efficiency and cost reduction gains.

Ontruck's AI-powered routing tool stands out for its unique features tailored to the specific needs of the transportation industry, offering advanced capabilities beyond static route optimisation. Dynamic route optimisation offers a compelling solution to rising costs and increasing customer expectations. To better understand this technology's transformative potential, contact us to schedule a demo. One of our experts will be happy to outline how dynamic route optimisation can streamline operations, cut costs, and elevate customer service to never-before-seen standards.