With the e-commerce industry expected to grow at higher rates over the next several years, the more successful parcel delivery companies continuously search for opportunities to efficiently reduce shipping prices, increase delivery speed, and increase service levels.
This PhD project aims to help parcel delivery companies thrive in such a competitive environment by guiding their strategic decisions as they devise a more integrated last mile network and fleet composition plan.
Parcel delivery companies compete on price, speed, and customer satisfaction to efficiently capture the expanding market in profitable urban areas. With an increasing demand for faster and seamless delivery experiences, parcel delivery providers need to focus not only on delivering on the right date but also at the right time in a cost-effective manner.
This research aims to support businesses involved in parcel delivery (e.g., e-retailers, postal operators, and third-party logistics providers) to optimise their logistics capital investment and improve service levels through more integrated last mile network design and fleet investment decisions. This is particularly relevant for parcel operators under price and service level threats from large and well-resourced competitors (e.g., Amazon and Alibaba).
Ultimately, making the right investment decision in logistics capacity becomes crucial for parcel operators to strive in the e-commerce era.
The objective of this project is to improve state-of-the-art mathematical optimisation models that can inform parcel operators on their logistics capacity planning in terms of network design and fleet composition decisions.
The main sub-objectives of this project are:
- Develop an integrated last-mile network design and fleet composition risk-averse optimisation model, which considers uncertainty in daily demand and the effects of seasonal peaks for long-term planning.
- Introduce the concept of demand segmentation in the optimisation model to distinguish the service-level requirements and operational complexities of each customer group.
- Develop a solution procedure, to derive near-optimal results for large instances within a reasonable model runtime. The three abovementioned sub-objectives fill an important research gap related to tackling uncertainties in last mile parcel distribution using practical robust solutions.
- Model and assess the financial impact of different failed delivery strategies (e.g., carding and reattempt deliveries) in the last mile network design and performance.
- Use traffic data from contemporary mapping technologies (e.g., Google Distance Matrix), to derive a data-driven extension of an augmented routing cost formulation to improve travel time prediction for local routes and the accuracy of the fleet composition plan.
More from iMOVE