Driving Sustainability through Route Optimisation and Delivery Management
Introduction
According to a report by Statista, the UK e-commerce market has been steadily growing, with sales amounting to £161 billion in 2022 and projected to reach £230 billion by 2025. Home delivery has played a significant role in this growth, with sales giants like Amazon, Ocado, and eBay leading the way. In fact, in 2020, the global online grocery market size was valued at $190.2 billion, a significant increase from $98.4 billion in 2019, according to ResearchAndMarkets.com.
The COVID-19 pandemic further accelerated the growth of home delivery, as households worldwide were forced to limit their physical movements. In the United States, for example, online spending for home delivery increased by 30% in 2020, according to a report by the National Retail Federation. This surge in demand contributed to the growth of the last-mile delivery sector.
The impact of the pandemic also caused many businesses to switch from B2B to B2C delivery models to survive. According to a survey by Accenture, 60% of businesses reported that the pandemic has accelerated their plans to invest in direct-to-consumer channels. This shift has further fueled the growth of last-mile delivery, as manufacturers and wholesalers now need to deliver directly to consumers.
Route optimisation
One of the primary bottlenecks in new delivery models is delivery planning. A good delivery process always starts with a delivery schedule or plan. With B2B organisations, those plans were polished within years to make them logical, simple to implement, and efficient. Fixed routes were well balanced to consider average volumes, business customer ranking, busy hours individual for each delivery location, and could be reused for many years.
With a change to B2C model delivery, planning became a daily routine as the volume of orders fluctuated, and the same household did not repeat the same order every day. Consequently, the same mechanism that historically worked for B2B planning was not helpful in the B2C world.
The complexity of route planning increases exponentially with the number of stops in a route. For instance, a route with four stops has only 24 potential options to consider, whereas a route with seven stops gives rise to 5,040 possible combinations. The number of possible combinations quickly grows into the millions as the number of stops increases. This phenomenon is known as factorial complexity, where the number of possible combinations increases much more rapidly than the number of stops. In fact, a route with 20 stops has 19 digits worth of possible combinations, while a route with 100 stops has a staggering 158 digits worth of possibilities. Below is an illustration of the planning complexity growing with a number of stops.
Number of Stops | Number of Sequence Combinations |
4 | 24 |
7 | 5,040 |
10 | 3,628,800 |
20 | 2.43 x 10¹⁸ |
100 | 9.33 x 10¹⁵⁷ |
150 | 5.70 x 10²⁶² |
Interestingly enough, the human brain is capable of processing basic routing challenges quite quickly when a dispatcher can imagine geography, locations, and routes between destinations. The simple example below shows a few points and the optimal sequence to follow: 1-2-3-4.
As geography gets richer around natural barriers like rivers or mountains, both dispatchers and drivers may easily become puzzled while defining the optimal sequence to follow. For example, a river between points 1-3 and 2-4 with bridges available to cross. Having those natural boundaries, the optimal delivery sequence will be 1-3-2-4.
One-way streets and congested areas will introduce further limitations to the delivery logic and potential stop sequence. The example below demonstrates one-way bridges only letting traffic in one direction. The optimal sequence will be 1-3-4-2 in this case.
There are more potential aspects that will make sequencing tasks even harder. The example below shows preferred delivery windows, which may be introduced by distribution contracts. This changes the delivery sequence to 1-4-3-2. Many other aspects may give the same sort of challenges, such as onboard-time limits for fresh products, driving time limits.
It's important to keep in mind that even with only 4 delivery locations, the complexity of route planning can quickly become overwhelming. This becomes especially true when dealing with realistic scenarios where drivers are making 20-100 stops a day. It simply isn't possible for humans or machines to consider every possible combination and select the optimal delivery scenario. To solve this problem, modern scheduling systems use heuristics and algorithms to quickly generate an initial delivery plan and then improve it over time with human input.
It's also important to note that the examples given so far have only looked at a single delivery route in isolation, while the average delivery fleet may have as many as 8 vehicles to manage. In reality, the first step in the sequencing process is breaking down the full list of orders between vehicles.
Routing and Scheduling solutions are a class of software systems available in the form of client-server or SaaS applications in your browser. These solutions specialise in cracking delivery complexity for small and large fleets, and can typically save between 10% to 30% of mileage while optimising delivery routes.
Their success can be attributed to several factors:
They have knowledge of all potential distances between planned points. Even before initiating the planning algorithm, they calculate all potential route combinations between points. For example, if you plan for 100 drop operations, they will start from 10,000 potential routes to calculate mileage and driving time between them.
They have detailed road networks loaded in their memory to navigate from A to B using the most efficient routes while respecting one-way streets and allowed roads by vehicle types.
They have a large hardware capacity running thousands of streams of calculations in parallel.
They use high-performance heuristics to look for optimal solutions using a powerful mathematical and statistical toolset.
They provide graphical planning interfaces that combine map and schedule views, which give dispatchers a visible and powerful way of amending the routes.
The implementation of route planning for a delivery fleet can significantly contribute to the sustainability of the planet. On average, each vehicle produces 28 kg of CO2 daily, based on a distance of 100 miles in the UK. For the typical SMB fleet of eight vehicles, a routing and scheduling solution could save up to 14 tons per year, with an average improvement in mileage of 23.2%. According to a study by the Aberdeen Group, companies that implemented routing and scheduling solutions saw an average reduction in travel time of 28.5%, a decrease in fuel usage of 23.2%, and an increase in workforce productivity of 15.2%.
Delivery efficiency
While route optimisation has a direct impact on the mileage of commercial fleets and environmental impact in general, everyday delivery efficiency would be another factor impacting the environment too.
Coming back to the migration from B2B to B2C delivery model, needless to say, delivery fleets were not prepared for a new reality. As larger vehicles usually deliver to 10-20 business locations had to explore new residential areas making 30-70 multi-stop drops a day.
Key factors which made the new delivery job extremely complex were:
Missing driver’s knowledge of the delivery area
Unknown geography of residential districts including one-way and narrow streets, traffic-heavy junctions
Generic postcodes not giving accurate delivery locations
Dynamic nature of B2C delivery with order volumes changing on a daily basis.
The factors above put drivers in a stressful mindset, causing aggressive driving, driving at higher speeds and excessive idling. According to the US Department of Energy, aggressive driving (including speeding, rapid acceleration, and hard braking) can lower gas mileage by up to 33% on the highway and 5% in the city, resulting in increased CO2 emissions. The UK Department for Transport estimates that idling can use up to 2.6 litres of fuel per hour and emit 6.3 kg of CO2, which can add up over the course of a day or a week of delivery driving.
A delivery management solution can help overcome the key challenges associated with B2C deliveries, such as the driver's lack of knowledge of the delivery area and the dynamic nature of B2C delivery orders. By providing real-time, turn-by-turn navigation, the delivery management solution can ensure drivers are taking the most efficient route to each delivery location, even in unfamiliar residential areas. This can reduce the likelihood of missed deliveries, improve customer satisfaction, and minimise the environmental impact of driving.
Moreover, a delivery management solution can provide accurate delivery locations by leveraging more precise address information and geolocation data. This can help ensure that drivers are able to find each delivery location quickly and efficiently, reducing the amount of time they spend driving around looking for the right address. The system can also dynamically adjust delivery schedules based on changes in order volumes or traffic conditions, helping drivers to complete their deliveries on time and reducing the amount of time they spend idling and emitting CO2.
In conclusion, the implementation of route planning and delivery management solutions can have a significant positive impact on the environment. Route optimization can reduce mileage by up to 30%, resulting in a reduction of up to 14 tons of CO2 emissions per year for a typical SMB fleet. Additionally, delivery management solutions can improve delivery efficiency, reducing the likelihood of missed deliveries, improving customer satisfaction, and minimising the environmental impact of driving. By adopting these solutions, businesses can not only improve their bottom line but also contribute to a more sustainable future.