The Evolution of Last Mile Logistics

With the rise of e-commerce and consumer behaviour evolving at a rapid pace, last mile logistics is becoming increasingly critical for efficient supply chain operations. Let’s discuss some of the forces driving the evolution in the last mile logistics, the inherent challenges, recent trends in the last mile delivery, and how technology and science are allowing for new advancements.

What are the forces driving the evolution?

There are three main factors driving the evolution in the last mile logistics space:

1. Urbanization: As more and more people move into cities, which is increasing the consumer density, the resultant constraints in capacity and infrastructure further complicate the overall shipment process and constantly change the way orders get fulfilled and delivered to end consumers.

2. Consumer Behaviour: As more people are getting used to on-demand consumption, where they get instant gratification by receiving their orders within hours, it’s forcing shippers to re-evaluate their last mile supply chain logistics and fulfilment policies.

3. E-commerce: The COVID-19 pandemic forced a lot of traditional brick and mortar shoppers to buy products online, which has changed the perception of the e-commerce experience for many. This is likely driving more traffic to e-commerce channels, which is causing many shippers to invest in auxiliary facilities and last mile deliveries to service new customer needs.

What are the key challenges?

Effective optimization of cost, service, and assets are the inherent challenges within last mile logistics. We know that urbanization brings uncertainty in supply and demand with traffic congestion, random disruptions such as road closures, diversions, accidents, and many other unplanned variables. These, along with constrained capacity and infrastructure (such as parking regulations, more frequent stops etc.), make modeling network design and routing for the last mile extremely complex. This is further complicated with the lead times running into hours, making it less flexible and cost efficient to service the customer within a short delivery time window.

How is the industry responding to this and what are the recent trends that we see?

With these variables driving more shippers towards last mile logistics, many have been moving towards an omni-channel strategy. The COVID-19 pandemic resulted in a surge in online demand, which further pushed this shift.

 In addition to traditional warehouse locations and auxiliary facilities in the urban areas, many shippers depending on service level and lead time are utilizing their existing brick and mortar retail stores to fulfil online orders through the storefront or backroom. As a result, last mile network design is moving from a traditional centralized approach to a more decentralized model.

Many carriers who have been reluctant to last mile runs in the past are now embracing its importance and strategically allocating their assets to accommodate. In urban areas, deliveries are now being carried out through multiple modes using various vehicle types, including crowd sourced, ranging from motorbikes to electric vans – both of which provide agility in the constrained urban infrastructure. However, this also poses challenges in integrating all sub-processes for full control and visibility into the delivery process to meet and exceed customers’ delivery expectations. Therefore, tracking and tracing becomes a critical part of last mile logistics. Shippers and carriers, especially after the ELD mandate, are now utilizing smart technology and sensors to track shipments at each step in the delivery process through better connectivity.

How can data science support last mile logistics?

Since connectivity and visibility are critical for the digitalization of the supply chain and logistics operations, data science plays an important role. Here are three ways:

1. Optimizing Solutions: As supply chains become more complex and difficult to model using traditional optimization methods (due to inherent uncertainty and unpredictability of last mile logistics), data science can help by modeling more accurate, cost-effective solutions using machine learning and a simulation-based stochastic approach.

2. Predicting Demand: Predictive modeling and AI/machine learning methods can help anticipate consumer demand and help move those SKUs to the local warehouses in advance. It can also help predict ETAs and service disruptions, which are essential in proactively notifying end customers and solving problems.

3. Dynamic Route Planning: With this data being readily available from several sensors and automated systems, it can be utilized to dynamically adjust routes based on moving demand, real time status of existing orders and routes, and lead times.

What are the value-added services that Coyote can provide in this space?

Coyote recently announced the launch of its dynamic route optimization program that aims to streamline supply chain operations and reduce uncertainty for carriers by maximizing the efficiency of their fleets and delivering load consistency through optimized weekly routing plans. This concept is now being further enhanced to solve for last mile route planning by introducing route learning, a data-driven approach, and adding more flexibility to take on dynamic orders with constrained windows.

To best position themselves to increase market share in the transportation industry, it’s important that shippers understand, evaluate, and optimize their last mile logistics operations.