This work presents a comprehensive framework to assist on-demand meal delivery platforms with the decision-making of delivery model choice and the planning of the micro-logistics centers within a mixed operational model. We consider two prevalent delivery models: (a) the owner-operator model operating with independent contractors (IC); and (b) the company-vehicle model with micro-logistics centers (CV). These centers serve as the place to accommodate company-owned vehicles. Assuming platforms have operated with independent contractors for some time, we aim to determine the necessity of micro-logistics centers, indicating whether to maintain the IC model or adopt a hybrid approach, and optimize the planning of necessary centers. We propose a mixed integer optimisation problem to minimize the total costs while considering the convenience for couriers. It combines strategic decisions for locating micro-logistics centers considering the dimensions of the centers (number, locations and vehicle stock) with operational considerations (the impact of couriers' distribution and shifts on repositioning company vehicles). For the CV model, two operational policies are considered: fixed coverage with return-to-origin requirement, and time-variant coverage with global redistribution considering the spatial-temporal variation of demand. Our findings suggest that diverse market conditions and operational approaches can lead to different strategies. We also applied the model for the city of Amsterdam, and it reveals that multiple centers are needed and the platform may invest in courier convenience by choosing center locations with great accessibility.

In response to pressing climate issues, Europe is transitioning towards more sustainable transportation. Despite the potential environmental benefits, train travel often loses out to other modes of transportation in long-distance journeys (300+ km). Based on previous research, transfers seem to have a great impact on the attractiveness of train journeys. This study investigates the effects of different transfer attributes on travel choice behaviour in long-distance leisure trips. Through a combination of a literature review and focus groups, key attributes influencing travellers' itinerary choices were identified and incorporated in a stated choice experiment. The experiment, involving 431 respondents, used a Latent Class Choice Model (LCCM) to analyze preferences and willingness to pay for improvements in transfer attributes. The study finds that travel cost and travel time are of the most importance. The number of transfers and transfer time are less critical but still impactful. The service level has a small impact. When the transfer occurs in the journey is insignificant. Four distinct classes of travellers were identified: Time-Valuing Eco-Advocates (44\%), Cost-Sensitive Train Travellers (25\%), Train Dislikers (18\%), and Transfer-Averse Train Sceptics (13\%). The results indicate that reducing travel costs and time should be prioritized over improving transfers to maximize the attractiveness of train travel. However, optimizing transfer time and minimizing the number of transfers will still enhance the appeal for specific traveller classes. By analysing the results, it was found that although direct train services are generally preferred, well-designed transfer options can improve market potential. This does however require significant improvements in the European railway network design and regulations.

E-commerce is rapidly growing and is expected to encompass a quarter of all global sales by 2025. This growth pressures e-commerce warehouses to enhance efficiency. A promising innovation is the Robotic Mobile Fulfilment System (RMFS), which optimises warehouse operations by using robots to manage storage and retrieval tasks, thus significantly improving productivity, speed and accuracy. This research focuses on how inventory allocation (slotting) decisions with RMFS can optimise operational performance. In particular, how the slotting decision of Stock Keeping Unit (SKU) distribution across movable storage racks (pods) based on SKU turnover can maximise order throughput rates and optimise operational performance. The research question guiding this study is: What is the optimal demand-based slotting decision to maximise the order throughput rate in a Robotic Mobile Fulfilment System? This question aims to provide insights into how different slotting configurations impact the efficiency and performance of ecommerce warehouses. The research approach is twofold. A general analysis is conducted to understand the impact of turnover-based slotting decisions using synthesised demand profiles derived from literature. This is followed by a detailed case study for Gall&Gall using demand profiles derived from real-world data to find specific optimal slotting configurations and validate the synthesised demand results. The methodology involves three main steps: determining demand configurations, generating slotting configurations with a mathematical model, and simulating these configurations to evaluate performance. Each demand configuration results in multiple slotting configurations, which are evaluated with the simulation to gain insights into the effect of slotting decisions on performance. The different demand profiles consist of total SKU quantity, total item quantity, and SKU classification into three classes (A, B and C) based on their item turnover. The different slotting configurations consist of different distributions of the three classes over the pods. These slotting configurations are obtained with a mathematical model that prioritises class distribution based on given weights. The simulation tool RawSim-O assesses the slotting configurations on key performance indicators such as total order throughput rate and the number of items picked from a pod in one go (pile-on). Key findings provide that pile-on and travel distance significantly affect the order throughput rate, with performance variations of up to 40 orders handled in 30 minutes. High performance often arises with configurations aiming for an equal number of items per pod across classes and maximising the number of pods for SKUs in class A. While synthetic demand profiles show high performance with class A distributed over the maximum number of pods or equal items per pod for all classes, the Gall&Gall demand profiles perform better with class B distributed over slightly more pods, indicating variability in optimal slotting approaches based on specific demand characteristics. Overall, turnover-based slotting decisions significantly impact order throughput rates in RMFS, and tailoring slotting configurations to specific demand characteristics is crucial for optimal operational efficiency. In addition to general slotting insights, this research developed a method that allows warehouses to input their specific demand characteristics and receive insights on optimal slotting approaches. Furthermore, the method enables the readjustment of warehouse-specific details, such as a warehouse’s unique layout, for extra applicability and realism, and allows the integration of additional decision problems, such as order batching and routing, to broaden the method’s scope. This supports warehouses with the design of a tailored, robust and effective slotting strategy for operational performance improvement

Private cars are inefficient in terms of land allocation. Car-free neighbourhoods offer a solution to this inefficiency by enforcing less car-centric urban design, expensive parking fees, and prioritising more sustainable and space-efficient modes of transportation. Carsharing services emerge as an alternative to private car ownership by providing vehicles available for short-term use. Notably, not much is known regarding carsharing services as an alternative mobility solution in car-free neighbourhoods. The present thesis investigates pricing strategies that could be implemented for business-to-consumer carsharing services in car-free residential neighbourhoods to align with the diverse needs and preferences of potential users. To this end, this thesis employs a case study of the car-free neighbourhood Merwede. Specifically, a cost-benefit analysis is conducted to estimate the potential value of carsharing in a car-free context. Emphasis is put on policies that may stimulate carsharing adoption and on determining equitable service pricing strategies.  To predict the modal split of future Merwede residents, datasets – containing mobility information for Utrecht residents clustered by income group – are used. To determine the influence of a service price change on carsharing demand, per-income-group price elasticities are estimated. This resulted in the following values for low-, middle-low-, middle-high-, and high-income groups, respectively: -0.8, -0.5, 0.4, and -0.1. Three stakeholders are considered in the analysis: carsharing service operators, users, and the Municipality of Utrecht. Results indicate that subsidies for lower-income residents in the form of a trip credit incentive (of 50 or 100 euros per month per user) yield positive total net present values and higher carsharing adoption rates. This thesis concludes by stating that while the generalisability of these findings may be limited, the CBA model synthesised for Merwede provides a case study from which future car-free neighbourhoods – aiming to implement carsharing – may learn.

The introduction of electric buses (EBs) is required in order to achieve global greenhouse emission goals. However, operating an electric bus fleet (EBF) requires careful consideration of both the mobility and energy systems in charging strategies. In literature and in real life many cases show careful consideration of only one of the systems and an oversimplification of the other, resulting in failed implementations. This paper focuses on the need for additional charging of EBs during the day using on-route opportunity charging during the on- and offboarding of passengers in residential neighbourhoods. This research aims to provide an integrated approach to bus service design, including both the mobility and energy systems. Additionally, this paper presents insights into the effect of traffic management on the feasibility of on-route opportunity charging. Two simulation tools are used for the integrated decision-making approach: SUMO and Gaia. The Simulation of Urban MObility (SUMO) software provides results regarding the energy consumption behaviour and charging needs of the EBs. Gaia provides results regarding the transformer load rate in the residential distribution grid. This paper proposes five strategies with a varying configuration of charging capacities, charging times and the introduction of traffic management via traffic priority. The five strategies are tested against three requirements: two from the mobility system and one from the energy system. The research presented uses a case study of bus line 36 in Rotterdam, the Netherlands. The results of this research show that residential neighbourhoods are often unable to handle chargers with the maximum capacity of the buses in the case study of 450 kW. A mid-range charger of 200 kW is possible in most cases. With a 200 kW charger, the on- and offboarding time of passengers is insufficient to complete daily operations. Therefore, additional charging time is necessary. Without the introduction of traffic priority, the timetable cannot be completed. With the introduction of traffic priority, travel time and energy consumption are reduced by 9.8% and 4.5%, respectively, and daily operations can be completed while satisfying the requirements of the mobility and energy systems. The results presented in this thesis suggest that an EB route design with opportunity charging is impossible without considering both mobility and energy systems simultaneously. From the mobility perspective, opportunity charging is feasible only if the charging infrastructure can meet the buses' energy needs without disrupting the schedule. From the energy perspective, it is essential that opportunity charging is implemented where and when sufficient charging capacity is available. The introduction of traffic priority allows for the satisfaction of all requirements and displays an example of the connection between mobility and energy systems.

The increasing parcel demand is resulting in increasing traffic congestion and emission problems in urban freight system. Crowdshipping service is an innovative logistics solution which envisions using the excess capacity in existing passenger transport to perform the delivery tasks. However, it may bring extra vehicle kilometres travelled due to detours and new trips generated. To address this challenge, most studies focus on designing and evaluating the crowdshipping service for all parcel demand, while one study points out the potentially larger benefits of crowdshipping service for outlier parcels.

Therefore, this study aims to investigate the transport impacts of crowdshipping service for outlier parcels, which are defined as the parcels with high environmental impacts. A case study is conducted in The Hague. First, the parcel carbon footprint is calculated to segregate the outlier parcels. Then, a public transport-based crowdshipping delivery scenario is proposed, with parcel lockers at train stations as the transfer points and train travellers as the potential occasional couriers. The simulation results show that outsourcing the outlier parcels to crowdshipping service is beneficial to the transport system and prioritising outlier parcels of logistics service providers with low market shares can achieve more savings in transport and higher service efficiency.

This research focuses on the complexities of integrating routing decisions, inventory management and pressure control in the context of the cryogenic logistics of Liquefied Natural Gas (LNG). This research aims to address these complexities by developing a planning method that considers the transportation and nitrogen cooling costs. The study evaluates the developed planning method through a case study at Rolande, which is a company that operates LNG and Bio-LNG refuelling stations in the Netherlands, Belgium and Germany.

This study considers three refrigeration methods for controlling the pressure level at refuelling stations. These are nitrogen cooling, offload cooling and logistic trailer cooling. The identified key performance indicators that are used to evaluate the develop planning method are: Total Costs, Total Transportation Costs, Total Nitrogen Cooling Costs and Cost per Kilogram. Furthermore, the key constraints and variables for the integrated control of the LNG logistics are identified.

The planning methodology was developed through three steps. First a Full Mixed-Integer Linear Programming (MILP) model was developed. Second, the Full-MILP model was refined to a Simplified MILP model. This Simplified MILP model was improved with the rolling horizon approach and a pre-solve process. The rolling horizon approach divided the planning horizon into smaller manageable time blocks. The pre-solve process identified the critical stations for each day to reduce the number of nodes in the network.

The proposed planning methodology was experimented in the case study that involved a network of 19 refuelling stations in the Netherlands and Belgium. The sensitivity analysis indicated that the model was sensitive to the vehicle capacity. Therefore, the pre-solve process was extended with determining the supply of LNG. The case study results revealed that the model is able to solve a seven-day planning horizon, while maintain the inventory and pressure levels within the specified bounds. However, the model can become computationally complex for days with high number of critical stations and cool vehicles.

This research contributes to inland LNG logistics by addressing the integration of routing, inventory and pressure management. Future studies should focus on a comparative analysis with heuristics, experimenting the feasibility and computational time with soft inventory and pressure bounds, and model a non-linear offload cooling effect to improve the realism of the model.

This study explores the potential role of e-scooters in supporting “first-mile” travel to train stations in the Netherlands, a critical stage in daily commuting for many Dutch travellers. In a nation where bicycles are a primary mode of transport—especially for trips under 5 kilometres—the introduction of e-scooters offers an alternative with similar benefits: environmental friendliness, cost-effectiveness, and ease of manoeuvrability. With their compact size and accessibility through sharing services, e-scooters can complement the existing bicycle culture, promising an appealing option for urban mobility.

The research focused on examining the role of e-scooters in “first-mile” travel through a stated preference choice experiment, given that e-scooters were newly legalised. A sample of 156 participants responded to six hypothetical travel scenarios, where they chose between familiar modes like walking, cycling, e-bikes, and e-scooters based on different factors such as carry-on ticket costs, parking times, and travel times. These scenarios were designed to reflect typical Dutch commuting choices while testing the appeal of e-scooters as a new transport option.

Analysis of the data using a Multinomial Logit (MNL) model yielded several insights into travel behaviour. There was a clear baseline preference for familiar modes, with participants showing a higher likelihood of choosing options like cycling or walking. Travel cost and time emerged as significant decision-making factors, with higher costs and longer times deterring mode choice. Notably, gender differences surfaced in this context, with males displaying greater tolerance for longer bike trips than females, indicating variations in time-cost sensitivities across demographics.

Experience with transport modes also influenced choices; prior e-scooter use was positively correlated with selecting this option again, suggesting that familiarity can increase comfort and confidence in choosing new modes. Established travel habits showed a strong impact on choice consistency, with individuals who typically walked or cycled to stations likely to maintain these habits, reinforcing the role of routine in travel preferences.

To illustrate the findings, a simulation modelled two scenarios. In the “Extreme Low” scenario, where e-scooter costs were low and parking time was minimal, walking was favoured for short distances (under 0.5 km), while cycling became dominant as distance increased. In contrast, the “Extreme High” scenario, with no fees and longer parking times, significantly boosted e-scooter attractiveness, particularly beyond 1.5 km, where it surpassed both walking and cycling. This outcome underscored the sensitivity of user preferences to economic factors, with e-scooters emerging as a highly competitive option when cost barriers were removed.

Overall, this study highlights the substantial potential for e-scooters to influence first-mile travel choices, particularly when costs are favourable and convenience is enhanced. By offering a viable alternative for medium-distance trips, e-scooters could reshape first-mile mobility in the Netherlands, complementing the well-established cycling culture and contributing to sustainable urban transport.

This project aims to relieve traffic pressure and enhance the parking experience for attendees during planned special events (PSEs). The objective is to develop an optimal strategy for efficiently allocating parking spaces during PSEs in parking lots.

PSEs, such as football games or large concerts, typically result in concentrated vehicle arrivals within a limited time period, leading to increased traffic flow, potential disruptions, elevated emissions, and safety concerns in nearby areas. By optimizing parking space allocation strategies in the parking lot, this project seeks to improve overall traffic management and relieve these challenges.

To achieve this, a linear programming (LP) algorithm and a simulation-based genetic algorithm (GA) are employed to search for the optimal solution. While the LP model offers computational efficiency, it has limitations in incorporating different route conditions. To address this, an agent-based simulation is constructed to depict the interaction and movement of vehicles within the parking lot. The simulation-based GA utilizes objective values derived from the simulation, providing a more comprehensive basis for finding the optimal solution. The allocation process considers factors such as parking lot layout, vehicle entry time step, and specific parking rules including road directions within the parking lot.

Results demonstrate that the optimal strategy obtained from the simulation-based GA outperforms comparison groups. The simulation-based GA showcases its ability to converge on the optimal solution within a large solution area. The optimal strategy saving time for all vehicles, particularly during periods of high demand. Effective parking is achieved by allocating parking spaces according to the arrival order and positioning vehicles on the left or right based on their arrival order and parking space location.

By employing these methods, this project offers a valuable contribution to the field of parking space allocation in the parking lot during PSEs, enhancing the overall parking experience for event attendees.

With the strong backing and advocacy from the EU for railway transport, it is crucial to focus on innovative and efficient technologies to maintain service quality. In daily train operations, traffic uncertainties lead to perturbations, which may result in two types of track conflicts: disturbances and disruptions.

In the current research, we aim to add and optimize timetable flexibility in traffic disturbance management for railway transport. A new definition is proposed for timetable flexibility. Timetable flexibility is defined as the ability of a timetable to be easily modified to withstand small disturbances and absorb delays, as well as to offer a larger solution space in the application of dispatching measures (retiming, reordering, rerouting) to solve larger disturbances without changing the given (re)scheduled timetable.
In order to minimize the deviation of the rescheduling plan from the rigid timetable, and maximize the timetable flexibility, the conflict resolution problem is modeled using an Alternative Graph (AG)-based Mixed Integer Linear Programming (MILP) model.

In order to investigate the impacts of different parameter inputs on timetable flexibility, a case study is conducted on a part of the Dutch railway network. To investigate the influencing factors of timetable flexibility, one illustrative application and two sensitivity analyses are conducted. Based on the results, practical implications on train dispatchers and signalers, as well as on railway passengers are concluded.

The impact of delays and disturbances in railway traffic can be mitigated by advanced rail traffic rescheduling models (RTRMs) which make use of mathematical optimization models. In the past several researches have been carried out on the effectiveness of an RTRM in reducing delay and improving punctuality. However, in most of these researches only one case study is used to test the RTRM. Still, it is unclear whether the results found for the application of an RTRM in one particular situation, are also valid for other situations (with different infrastructural and operational characteristics).
With this research, the impact of different infrastructure layouts and traffic patterns on the effectiveness of rail traffic rescheduling models is investigated. It provides insight into whether the benefit of an RTRM depends on the infrastructure and timetable in the area where it is applied.
For this purpose, an evaluation framework has been developed in which an RTRM can be tested using different infrastructure, operational and disturbance scenarios. In this framework, the RTRM is used to generate a real-time traffic plan for each scenario. These traffic plans are compared with the traffic plans of simple dispatching rules, that can be used in practice by dispatchers. For this comparison, KPIs such as the sum of consecutive delay (amount of delay that propagates within the network) and punctuality are used. This framework is applied to an alternative graph-based RTRM, which is formulated as a MILP (mixed integer linear programming).
The results show that the improvement an RTRM can offer over simple dispatching rules, varies per infrastructure layout and traffic pattern. For some infrastructure and operational scenarios, the simple dispatching rules perform as well as the RTRM, which means that for these situations, implementing an advanced RTRM does have much added value. A trend has been observed that the effectiveness of the RTRM increases as more control options are available.

With the number of passenger cars increasing worldwide, shared modes are being introduced in addition to public transport and active modes to reduce private car usage. The concept of mobility hubs was introduced to improve the accessibility of the different available transport modes. At a mobility hub, all sorts of mobility come together, and additional facilities such as pick-up points and restaurants are added to increase the attractiveness of the hub. Based on stated-preference surveys and expert interview reports, it was identified that mobility hubs could change people's travel behaviour in the region where it is located. However, these studies only provided insights and were insufficient to assess the mobility hub’s impact on the transport network. The parking demand in residential areas is also increasing pressure on residents to switch to different modes, and these hubs could play a key role in providing alternate transport options. As a result, this research aims to study the impact of mobility hubs in a residential area on the transport network in which it is placed. This study identifies the typology of mobility hub suitable for a residential area and the components required at the hub. This study also proposes a generalized methodology for introducing mobility hubs with shared modes in aggregate transport models. A case study is performed on the Delft transport network in OmniTRANS based on the proposed methodology to understand the impacts the mobility hubs can cause. In this case study, only the shared modes are assumed to be available at the hubs, and the additional facilities are not considered. Based on the simulation results, it was found that the travel behaviour changes majorly in the residential regions where the mobility hubs have been modelled. The private car trips were found to decrease marginally, but the total number of car trips, including both private and shared cars, increased. This research recommends further research on people's behaviour in the region in the form of pilot studies before modelling for simulation as people’s behaviour varies from region to region.  

Crowding in trains during rush hour is known to cause discomfort. After the outbreak of COVID-19 pandemic, crowding has also been highlighted as a risk factor for catching Acute Respiratory Infections (ARIs) such as COVID-19 which has affected the demand of public transport. Several countries, including the Netherlands, have differential fare systems for peak and off-peak travel, however, the problem of overcrowding in trains is still prevalent and is expected to cause more disutility than before the pandemic. To reduce peak hour rush, change in departure time has proven to be an effective measure. In this research first a review of previous experiments related to valuation of crowding and departure time change is performed. And then an exploratory study based on a stated choice experiment is conducted to understand the extent to which people can be motivated to change departure time to avoid crowded trains during rush hours by offering them real time information on on-board crowding level and a discount on train fare. Unlike previous studies, the respondents are segregated into two groups before the start of choice experiment based on their indicated preference to schedule delay early or late. To study the change in travel behavior in the pandemic time, context of different vaccination stages is provided in the choice experiment. Background information collected in the experiment is broadly categorised as: socio-demographics, travel and work related factors, and attitude towards health and COVID-19. After the responses are analysed it is found that the coefficients obtained for main attributes are highly significant, and in line with previous research. When most of the people are vaccinated in the Netherlands, they may become less averse to on-board crowding. The research also indicates that certain groups of people can be motivated to change their departure time if real-time crowding information is provided to them. Few others can be motivated by offering incentives. However it should be noticed that to allow people to change departure time, policies such as flexible work hours and staggered commute are required in workplaces.

E-grocers are grocery markets which offer their assortment online and deliver a customer's groceries at home. For an e-grocer the quality of their delivery service is a crucial asset to create and maintain a loyal customer-base. Because the online share of the grocery market has grown rapidly over the last decade, there is an urgent need for e-grocer specific routing models. In parallel, the research community has made rapid developments in the fields of stochastic and dynamic routing models during the past decade. Those studies suggest that the use of stochastic and dynamic elements improves the performance of routing models for a large variety of applications. However, it is yet unresearched how these stochastic and dynamic routing models can be used to improve the last-mile delivery for e-grocers specifically. The intended outcome of this research is the formulation of a promising modelling approach to use stochastic and dynamic routing model elements to improve the performance of an e-grocer's routing model. In this thesis, first the objectives and requirements for an e-grocer's routing model are determined in more detail by means of a case-study at Dutch e-grocer Picnic. The strict requirements for an e-grocer's routing model are: \begin{enumerate*}[label=(\arabic*)] \item Offer customers a free one-hour delivery time window when they place an order. \item Offer free communication of a 20-minute delivery time window at the morning of the delivery. \item Calculation of the combinations of customers in a trip has to be completed within 45 minutes. \item The trip planning has to be calculated within 5 hours. \end{enumerate*} When these requirements are met, the objectives for the routing model are to maximize the on-time delivery performance and to minimize the operational costs. The on-time delivery performance is quantified by means of the on-time delivery rate and the rate of extreme lates (\geq 15 minutes late). The operational costs are quantified by means of the average time spent per delivery. Next, scientific literature on the topic of dynamic and stochastic routing models is studied. Based on the findings in this literature study, it is suggested to make a distinction between ``regular” deliveries and ``flexible” deliveries. Regular deliveries have to be delivered within a 20-minute delivery time window. Flexible deliveries allow for larger delivery time windows. Once the trip has departed, the presence of these flexible deliveries allows for dynamic re-optimization of the customer sequence in that trip. This strategy has the potential to improve the on-time delivery performance. Moreover, this approach does not incur large additional operational costs because the number of vehicles used for delivery remains the same. At e-grocers, limited computation time is available for the calculation of the combinations of customers in a trip. However, plenty of time is available to determine the exact trip planning. Therefore, the traditional vehicle routing problem is split into two separate problems which are solved one after the other: The customer-trip assignment problem and the sequencing problem. Then, different approaches for implementing the concept of flexible deliveries in an e-grocer's routing model are designed. Three different sequencing models are formulated to get insights in the benefits of a stochastic routing model: A deterministic sequencing model which resembles the current operations at Picnic (benchmark model), a stochastic sequencing model which uses a-priori simulation to choose the best solution out of a set of good solutions (simulation-based model), and a deterministic sequencing model which positions the flexible deliveries in the customer sequence of a trip in such a way that the effectiveness of re-optimization is maximized (heuristics-based model). In order to study the added value of a dynamic routing model, a deterministic re-optimization model is designed. In order to quantify the performance of a routing model which makes use of flexible deliveries and compare the different model variants, four different routing model configurations are investigated by means of a computational experiment. A configuration consists of a sequencing model and, in some cases, a re-optimization model. The computational experiment uses real historic test instances from e-grocer Picnic to evaluate the performance of the configurations. In addition, historic data from Picnic is used to simulate realisations of planned trip times. Two different types of customer-trip assignments are investigated: customer-trip assignments for a small vehicle and for a large vehicle. Moreover, two different sizes of the flexible delivery windows are studied: 60 minutes and 75 minutes. From the experimental results it can be concluded that the configurations including a re-optimization model outperform the static and deterministic benchmark configuration in terms of on-time delivery performance. However, re-optimization comes at the price of an increased average time spent per delivery. This can be explained by the fact that the re-optimization model optimizes the on-time delivery performance instead of the trip duration, resulting in larger average travel times. The effects of re-optimization become significant when 10\% or more of all deliveries is flexible. For this percentage of flexible deliveries, the number of late deliveries can be reduced by up to 18\% and the number of extreme lates can be reduced by up to 27\%, depending on the type of vehicle and the size of the flexible delivery windows. The simulation-based sequencing model proves to be effective without re-optimization for the large vehicle type. For the small vehicle type it needs the re-optimization model to significantly outperform the static and deterministic benchmark configuration. When the re-optimization model is included in the configuration, the simulation-based and heuristics-based sequencing models show similar performance in terms of on-time delivery rate. The configuration with the simulation-based approach results in a lower average time spent per delivery. However, the heuristics-based approach results in fewer extreme lates. In order to decide on the optimal routing model configuration for a specific e-grocer, the relative importance of the on-time delivery performance and the operational costs have to be specified. If an e-grocer is very cost-sensitive, the configuration consisting of the simulation-based sequencing model and the real-time re-optimization model would be optimal. However, if an e-grocer has more financial possibilities for an increase in operational costs the configuration including the heuristics-based sequencing model would be a better fit. Extending the flexible delivery windows results in a significant improvement of the on-time delivery performance of the configurations. However, the increased number of successful re-optimizations comes at the expense of an increased average time spent per delivery. The added value of extended flexible delivery windows is largest for the small vehicle type. When comparing the two types of vehicles studied, it becomes evident that the use of stochastic and dynamic routing model elements is more effective for the large vehicle type than for the small vehicle type.

The petrol distribution problem is well-known in the literature as the Petrol Station Replenishment Problem (PSRP). This research concerns a "rich" version of the multi-period PSRP with real-life characteristics, to represent the complexity of the practical problem. The planning is optimised over multiple days, since stations do not need to be replenished every day. Mixed-Integer Linear Programming (MILP) models and a decomposition heuristic are proposed as planning methods, which are evaluated with a case study based on a real-life petrol distributor. Variants to these MILP models are proposed for the situations where the inventory is allowed to drop below the safety stock level, where inventory levels need to be minimised and where the service time depends on the delivery quantity. Moreover, to determine delivery quantities, the heuristic uses the new introduced simultaneous dry run inventory policy. An improvement procedure is applied to improve the initial heuristic solution. A commercial solver is able to find solutions for instances with up to 20 stations and 7 days for the MILP models. Exact solutions are found for instances up to 10 stations and 5 days. A heuristic solution was found for the full case study of 59 stations and 14 days, within the time limit of two hours.

Urban freight transport is a significant cause of CO2 emissions. This creates the need to implement CO2 mitigation measures for freight transport in cities. The International Transport Forum (ITF) is currently working on an EU-funded programme, the Decarbonising Transport initiative, to assess CO2 mitigation measures. However, there is no model yet that is suitable to analyse urban freight transport on a European level. Therefore, the research question of this thesis is: How can a model for urban freight transport in the urban region of Rotterdam (Groot-Rijnmond) be created, that allows for estimation of CO2 impacts of policy measures for decarbonisation and that can be transferred to other European urban areas? To answer this question, a literature study has been done to set the context of the project. This was followed by an identification of available data sources for European freight transport and the creation of a transferable modelling methodology. Based on the data and the established methodology, a model has been estimated. With this model, seven policy scenarios have been tested.

The objective of this research is to understand what smart charging can bring business and society at large. In close collaboration with the largest fleet operator in the world and a commercial aggregator, the impact of smart charging on both cost reduction and carbon reduction was simulated for an electric vehicle (EV) fleet in 2018. The simulation is designed to quantify the cost reduction of EV smart charging in The Netherlands as realistic as possible. Besides cost reduction quantification, the objective is to create a better understanding of what variables influence smart charging cost reduction. This is done via a statistical analysis of the smart charging simulations output. Furthermore, this research also has the objective find the direct impact of smart charging on carbon intensity of the electricity used for personal transport.
In this research, an EV aggregators perspective is leading. The EV aggregators could utilizing available flexibility in an EV fleet to deliver flexibility services. The strategy chosen to simulate is based on day ahead market optimization and passive balancing on the imbalance market. The EV fleet is assumed to be an isolated portfolio handled by the balance responsible party (BRP). The average synthetic load profile over 2018 was €41,56 per MWh and this is used as benchmark to quantify the smart charging savings in the simulations.
Different smart charging simulation set-up scenarios are designed and executed. In all simulations a real-world charging data set with 300.000 historic charging sessions was used. For each session, a new smart charging profile is determined by the optimization algorithms. The session price and session carbon intensity is calculated for both the smart charging scenarios as the business-as-usual scenarios. To compare the results of the different smart charging set-up scenarios, the average session price of all session in that simulations is used. At the same time, the smart charging savings is calculated based on the defined benchmark. The findings within this thesis support the conclusion that the used smart charging algorithms work properly and could decrease the electricity purchase price in The Netherlands. Additionally we found that the carbon intensity of the charged electricity during the smart charging schedule decreases compared to a business as usual scenario. This is a direct result of a correlation between the carbon intensity in the grid and day ahead prices in The Netherlands. EV aggregators are able to add flexibility to the demand side of the electricity system by means of smart charging, if a strong price incentive is provided. If stakeholders across the mobility and the energy sector work together, a real-world commercial implementation based on the price incentives on day ahead market and imbalance market in The Netherlands is possible. In the statistical analysis, multiple regression models show a linear relation between three independent variables (the session duration, session volume and maximum power of the charge point) and two dependent variables (the average session purchase price and savings per session). The key insights from the models empowered three main recommendations to EV aggregators to optimize the smart charging savings in the future: 1) Encourage longer session length. 2) Encourage regular overnight charging sessions behaviour, independent from the charging needs. 3) Stimulate access to high charging power. The data showed compelling differences between the 20% BEVs and the 80% PHEVs and their results were separated accordingly in this research. In all simulation set-up scenarios are the PHEVs outperforming the BEVs in terms of a lower average session price and higher cost reduction. If the smart charging strategy is executed as proposed in this thesis, the EV aggregator is exposed to the day ahead market and imbalance settlements for its portfolio. The EV aggregator is able to decrease the electricity purchase price, while acting as BRP. The exposure to the markets brings significant risk. Collaboration with an electricity supplier or BRP could potentially increase the smart charging savings for the EV aggregator. Furthermore, other revenue streams to utilize flexibility could be investigated. If stacking different flexibility strategies is possible, it could increase the smart charging value in the future.

In the transportation planning process, the Four Step Model (FSM) is used to define the needs and requirements of the transportation system within a city or a region. Despite its wide use, the model is focused on vehicular trips and fails to represent the demand of walking activity. The limited work that has been performed to address the misrepresentation of walking in the context of the FSM, still does not address the last step of the four step model, the assignment of the walking trips to the network. Stemming from this literature gap, and the need to enhance the role of walking activity within the transportation modeling, the main question of this thesis is to develop a method for the assignment of the walking trips to the street network. This research suggest a methodology to model large continuous space and walking trips on it. Thus, the focus is on modeling walking space taking into account the pedestrian scale for the model. To implement that, the size of the spatial analysis zones of the FSM is adjusted to walking scale. Next, continuous walking space is defined by the BGT spatial dataset that offers information for the land covers. In order to model continuous walking space two space discretization techniques are compared; first, Constrained Delaunay Triangulation (CDT) is applied on the walking area polygon, second, a regular quadrilateral grid is overlaid with the walking area. Walking activity is modeled on the network that is formed with the dual graph of each discretization method. In a subsequent step, the trip injection is performed with zone centroids and connectors which were created with the point clustering algorithm DBSCAN having as input points the graph nodes. Finally, A* algorithm is used for deterministic an all-or-nothing network assignment. The results of the applied methodology show that addressing walking within the FSM requires a finer grained approach due to the more localized scale of walking activity. Walking space needs to be defined as a continuous surface. In order to model such surface at the scale of the FSM, a discretization method is required to handle continuous space at large scale. Comparing the result of the two discretization techniques, polygon triangulation and regular grid, the triangulation dual graph results to the ’skeleton’ of the polygon while the regular grid dual graph offers a more dense and homogeneous polygon representation. Additionally, in the trip injection process, the nodes of the regular grid perform better than the triangulation graph nodes. Overall, throughout the whole implementation, the performance of using a regular grid, outweighs the polygon triangulation as discretization method.

In this paper, we study the effect of a low parking requirement in combination with dedicated carsharing at a residential building on car ownership and use of carsharing. This paper aims to develop a conceptual model to describe these effects. Comparable to customary theories in transport geography, this conceptual model includes socio-demographic factors, attitudes, residential relocation and the built environment. Besides, the interests, goals and experiences of stakeholders that include these combinations are analysed. The analysis is based on a series of 12 semi-structured interviews with carsharing providers, municipalities and developers in the Netherlands. Using two examples of residential buildings that include a low parking requirement and carsharing, this article examines to what extent parking requirement and offering carsharing leads to a reduction of car ownership and use of carsharing compared to control cases. Based on online surveys (n=18/68) carried out, descriptive analyses and binary analysis were performed.
The findings show that the stakeholders have the same goals. However, they have different experiences regarding car use and car ownership. The effects on car ownership and use of carsharing may also be influenced by other factors that are described in the conceptual model. Lastly, the findings of the differences between the selected and the control cases, show that it often appears that households in the selected cases did not have a car beforehand. Moreover, offering a shared car does not have a clear influence on its use. However, the use of the shared car and the ownership of the shared car may be explained by other factors. Moreover, the differences are not statistically significant.

Vehicle automation has the potential to disrupt the status quo of urban transportation, because it adds a new mode of transportation by taking away the task of the driver. The projections of estimations of the impacts of this new technology are based on many uncertainties and are thus largely unknown. A literature review showed that a wide number of effects is possible with automated driving and that no straightforward method to assess future impacts of this new technology exists yet. This thesis developed a method that provides insights into the impacts of vehicle automation in urban areas, without preconceived ideas about the impacts of the different scenarios. Existing knowledge from literature, transportation and land use data, and sociodemographic information were combined in a geographically disaggregated
System Dynamic model. This model explored the effects of vehicle automation on the performance of the transportation and spatial system of the case-study city of Copenhagen, Denmark. Different model runs provided insight in the possible range of outcomes. Considerable problems may arise in the transportation network with the introduction of automated driving because, using the car might become very attractive. A city’s land use does, however, not change as much as many could expect. The causes of (un)desirable outcomes
were identified with the Patient Rule Induction Method. The ranges of uncertainties in the value of time in an automated vehicle and in the level of adoption of car-sharing were found to influence desirable versus undesirable futures the most. Mitigating measures should focus on these scenarios to prepare for a future with automated driving.