https://hdl.handle.net/1721.1/123316 2026-04-04T18:57:56Z 2026-04-04T18:57:56Z Zinnenlauf, Philipp Pina-Pardo, Juan C Winkenbach, Matthias https://hdl.handle.net/1721.1/155674.2 2025-04-07T09:03:35Z 2024-07-03T00:00:00Z

The Value of Demand Forecasting in Stochastic Last-Mile Fleet Sizing and Composition Planning Zinnenlauf, Philipp; Pina-Pardo, Juan C; Winkenbach, Matthias Customer demand constitutes a crucial source of uncertainty in designing and operating complex and costly urban last-mile distribution operations. To mitigate associated risks, companies are diversifying their last-mile delivery options, exploring new vehicle types, and engaging in varied contracting schemes, encompassing vehicle rentals and spot market capacity utilization. We introduce a sequential learning and optimization problem integrating demand forecasting into a tactical last-mile fleet composition problem under uncertainty. Specifically, we propose a novel forecasting infrastructure and several machine learning models to predict customer demand in the medium-term future with high granularity. These forecasting results are then integrated into a two-stage stochastic program to derive cost-optimal fleet compositions. A real-world case study focusing on an e-commerce retailer in São Paulo, Brazil, reveals the economic viability of stochastic fleet composition planning informed by highly accurate demand forecasts. Our results show that accurate demand forecasts enable e-commerce retailers to make cost-minimizing tactical decisions about the size, vehicle type, and governance structure of the rented vehicle fleet. Furthermore, our framework underlines the importance of implementing integrated decisions, where a fleet composition design is interlinked with forecasting methods to mitigate uncertainties.

2024-07-03T00:00:00Z Zinnenlauf, Philipp Pina-Pardo, Juan C Winkenbach, Matthias https://hdl.handle.net/1721.1/155674 2025-01-23T13:45:24Z 2024-07-03T00:00:00Z

The Value of Demand Forecasting in Stochastic Last-Mile Fleet Sizing and Composition Planning Zinnenlauf, Philipp; Pina-Pardo, Juan C; Winkenbach, Matthias Customer demand constitutes a crucial source of uncertainty in designing and operating complex and costly urban last-mile distribution operations. To mitigate associated risks, companies are diversifying their last-mile delivery options, exploring new vehicle types, and engaging in varied contracting schemes, encompassing vehicle rentals and spot market capacity utilization. We introduce a sequential learning and optimization problem integrating demand forecasting into a tactical last-mile fleet composition problem under uncertainty. Specifically, we propose a novel forecasting infrastructure and several machine learning models to predict customer demand in the medium-term future with high granularity. These forecasting results are then integrated into a two-stage stochastic program to derive cost-optimal fleet compositions. A real-world case study focusing on an e-commerce retailer in São Paulo, Brazil, reveals the economic viability of stochastic fleet composition planning informed by highly accurate demand forecasts. Our results show that accurate demand forecasts enable e-commerce retailers to make cost-minimizing tactical decisions about the size, vehicle type, and governance structure of the rented vehicle fleet. Furthermore, our framework underlines the importance of implementing integrated decisions, where a fleet composition design is interlinked with forecasting methods to mitigate uncertainties.

2024-07-03T00:00:00Z Parks, Steven Winkenbach, Matthias https://hdl.handle.net/1721.1/152374 2023-10-05T03:38:33Z 2023-10-01T00:00:00Z

Estimating the Traffic Congestion Footprint of Retail E-Commerce Parks, Steven; Winkenbach, Matthias The last decade has seen pivotal changes in the "last mile" of logistics: online order volumes are growing, customers are expecting deliveries with quicker turnarounds, and new technologies such as drones and electric vehicles are reshaping suppliers' fulfillment strategies. This study estimates the effects of e-commerce logistics on urban mobility, seeking to quantify how efforts by e-commerce suppliers and carriers to meet shifting consumer demands impacts traffic congestion as experienced by city drivers. Because the impacts to traffic caused specifically by e-commerce operations is not directly observable, we estimate these effects using traffic simulation models. We find that the impact of last-mile e-commerce activities to traffic congestion depends heavily on changes to people's travel behavior as they purchase more goods online. If people make fewer daily shopping trips to brick-and-mortar stores, traffic congestion will improve. If people complement online shopping with in-store shopping, traffic congestion will degrade. To mitigate potential negative traffic impacts, we recommend carriers reform their delivery networks to limit the distances delivery vehicles must travel within cities. Introducing micro-fulfillment centers close to end customers, serving secure lockers where consumers can pick-up packages, transitioning to alternative delivery vehicles with smaller traffic footprints, and consolidating packages onto fewer delivery vehicles with larger cargo capacities can all help reduce the traffic congestion induced by last-mile operations.

2023-10-01T00:00:00Z Lehmann, Jonas Winkenbach, Matthias Janjevic, Milena https://hdl.handle.net/1721.1/151744 2023-08-09T03:44:40Z 2023-08-08T00:00:00Z

Operational and tactical levers to reduce carbon emissions in temperature-sensitive freight transportation for pharmaceuticals Lehmann, Jonas; Winkenbach, Matthias; Janjevic, Milena Carbon emissions are surging worldwide, with mounting political targets and shareholder pressure driving decarbonization efforts across all sectors. The temperature-sensitive pharmaceutical industry is particularly carbon-intensive, partly due to its reliance on carbon-intensive air transport. As pharmaceutical companies predominantly outsource transportation services, they have become increasingly interested in understanding how to minimize emissions from shipping their goods. Consequently, we introduce a carbon estimation and allocation tool tailored to cold-chain logistics for pharmaceuticals, addressing several unique aspects not considered in previous studies, such as incorporating network-level emissions for cooling container provision and a more granular approach for estimating transportation emissions. Applying the tool to a series of case studies, we discover that cooling containers with a lower weight and a higher ratio of usable interior volume to outer container volume have a low r carbon footprint, primarily due to reduced emissions from airfreight. In addition, the optimal container choice depends on the shipment size to maximize the container fill rate. Moreover, the provisioning of cooling containers to alleviate network imbalances via maritime or airfreight should be thoroughly assessed, as it may significantly increase a shipment’s carbon emissions.

2023-08-08T00:00:00Z Gaba, Farri T. Winkenbach, Matthias https://hdl.handle.net/1721.1/145726 2022-10-07T03:00:58Z 2022-10-03T00:00:00Z

Regulatory Implications for Unmanned Aerial Vehicles in Last-Mile Delivery Gaba, Farri T.; Winkenbach, Matthias This report examines the legal, regulatory, and societal barriers facing unmanned aerial vehicles (UAVs) as they are deployed for last-mile delivery with a particular focus on deployment in the United States (U.S.). The status quo for current legal and regulatory restrictions is first explored followed by a discussion on miscellaneous regulatory issues and areas of uncertainty that are likely to challenge unmanned aerial vehicles for last-mile delivery (UAV-LMD) deployment. This is followed by an evaluation of potential societal barriers imposed on UAV-LMD and the likely regulatory pathways through which these barriers will manifest. This report offers a nominal set of restrictions likely to limit operators and their fulfillment network planning and strategic design.

2022-10-03T00:00:00Z Winkenbach, Matthias Parks, Steven Noszek, Joseph https://hdl.handle.net/1721.1/131235 2021-10-09T05:02:38Z 2021-09-02T00:00:00Z

Technical Proceedings of the Amazon Last Mile Routing Research Challenge Winkenbach, Matthias; Parks, Steven; Noszek, Joseph Winkenbach, Matthias; Parks, Steven; Noszek, Joesph In the spring of 2021, Amazon hosted the Last Mile Routing Research Challenge with support from the MIT Center for Transportation & Logistics. The goal of the challenge was to encourage participants to develop innovative approaches to produce solutions to a real-world route sequencing problem, leveraging a large set of real operational data released by Amazon. All finalists were asked to submit short technical papers to document their methodological approach and its expected performance. This proceedings document provides a consolidated overview of the papers submitted by the participants.

2021-09-02T00:00:00Z