Datasets supporting PhD dissertation 'Autonomous goods vehicles: implications for fleet operating models'

This repository includes datasets and programmes supporting the PhD dissertation 'Autonomous goods vehicles: implications for fleet operating models'. The contents are described as follows: Chp 3: Selection of vehicle size and extent of multi-drop deliveries for autonomous goods vehicles Folder contains all sets of VRP analyses described in the PhD. Analyses include full datasets, radial and random data subsets for 'Customer Deliveries' and 'Intermediate Distribution' case studies. Each folder consists of the following files: DistMatrix.txt: This is a matrix of distances between each destination in the analysis in kilometres FVOT.txt: This a small table listing the different values of FVOT (Freight Value of Time) LocDemand.txt: This is a table listing the reference numbers of the destinations in the dataset, the quantity of cargo required to be delivered to each destination and the coordinates of the destination. To preserve commercial confidentiality, coordinates have been removed for the uploaded datasets. In the 'Intermediate Distribution' cases, cargo loads are created in multiples of 12 cages with dummy location numbers used to differentiate between each multiple of 12 cages. VehParams.txt: This is a table of vehicle parameters including hourly driver and vehicle cost (in £/hr), mass capacity of the vehicle (in tonnes), volume capacity (in units of cargo, e.g. crates for LGVs and cages for rigid/tractor-trailer), mass per unit volume, fuel consumption when the vehicle is empty and full respectively. LWFMVRP.py: This is the Python program which runs the optimization analysis using Gurobi commercial solver. Output.xlsx: This is the excel output of the analysis and includes a summary as well as output matrices for the best solution discovered by the analysis. Chp 4: Opportunities for off peak deployment Folder contains all sets of VSP analyses described in the PhD. Analyses include the multi-destination case study as well as 30 journeys of the 'High congestion' example and of the 'Low congestion' example. Analyses are labelled as per the PhD. Each folder consists of the following files: Aik.txt: This is a matrix of one-way journey duration in hours for each destination for each hour of the day (Tk=0:23) Cdik.txt: This is a matrix of return journey driver time cost in £ for each destination for each hour of the day (Tk=0:23) Di.txt: This is a list of 'latest delivery' restrictions for each destination. Fc.txt: This is a matrix of return journey fuel consumption in litres for each destination for each hour of the day (Tk=0:23) Pik.txt: This is a matrix of return journey duration in hours for each destination for each hour of the day (Tk=0:23) Ri.txt: This is a list of 'earliest arrival' restrictions for delivery for each destination. Tk.txt: This is a list of the hours of the day for which the data pertains to (Tk=0:23) TDFSVSP.py: This is the Python program which runs the optimization analysis using Gurobi Commercial solver. Output.xlsx: This is the excel output of the analysis and includes a summary as well as output matrices for the best solution discovered by the analysis. Chp 5: Opportunities for integrated operating model changes Folder contains all sets of VSP analyses described in the PhD. Analyses include the multi-destination case study as well as 30 journeys of the 'High congestion' example and of the 'Low congestion' example. Analyses are labelled as per the PhD. RB=Baseline speed strategy. S1: Speed strategy 1 (80km/h max). S2: Speed strategy 2 (70km/h max). Free: Speed strategy is modelled as a decision variable. Each folder, other than the analyses denoted 'Free' consists of the following files: Aik.txt: This is a matrix of one-way journey duration in hours for each destination for each hour of the day (Tk=0:23) Cdik.txt: This is a matrix of return journey driver time cost in £ for each destination for each hour of the day (Tk=0:23) Di.txt: This is a list of 'latest delivery' restrictions for each destination. Fc.txt: This is a matrix of return journey fuel consumption in litres for each destination for each hour of the day (Tk=0:23) Pik.txt: This is a matrix of return journey duration in hours for each destination for each hour of the day (Tk=0:23) Ri.txt: This is a list of 'earliest arrival' restrictions for delivery for each destination. Tk.txt: This is a list of the hours of the day for which the data pertains to (Tk=0:23) TDFSVSP.py: This is the Python program which runs the optimization analysis using Gurobi Commercial solver. Output.xlsx: This is the excel output of the analysis and includes a summary as well as output matrices for the best solution discovered by the analysis. In the folders denoted 'Free', the above files aik, cdik, fc and pik end with numbers 0, 1 or 2. These correspond to the respective speed strategies (base, S1 and S2, as described above)

本仓库包含支持博士学位论文《自主货运车辆：车队运营模式的影响》的数据集与程序，具体内容说明如下：
第3章：自主货运车辆的车型选择与多站点配送范围
本文件夹包含博士论文中提及的全部车辆路径问题（Vehicle Routing Problem, VRP）分析数据集。分析涵盖“客户配送”与“中间配送”两类案例研究的完整数据集、径向与随机数据子集。每个文件夹包含以下文件：
DistMatrix.txt：本文件为分析中各目的地间的距离矩阵，单位为千米。
FVOT.txt：本文件为一张小型表格，列出了货运时间价值（Freight Value of Time, FVOT）的不同取值。
LocDemand.txt：本文件为一张表格，列明了数据集中各目的地的编号、各目的地所需交付的货物量，以及目的地坐标。为保护商业机密，上传的数据集中已移除坐标信息。在“中间配送”案例中，货物装载量以12个笼箱为单位递增，同时使用虚拟位置编号区分每一组12个笼箱的装载量。
VehParams.txt：本文件为车辆参数表格，包含驾驶员与车辆的每小时成本（单位：英镑/小时）、车辆质量载荷上限（单位：吨）、体积载荷上限（以货物单位计，例如轻型货车（Light Goods Vehicle, LGV）的货箱，以及重型货车/牵引挂车的笼箱）、单位体积货物质量、车辆空载与满载时的燃油消耗量。
LWFMVRP.py：本文件为Python程序，使用Gurobi商业求解器运行优化分析。
Output.xlsx：本文件为分析的Excel输出结果，包含分析得出的最优解的汇总信息与输出矩阵。
第4章：非高峰时段部署的机遇
本文件夹包含博士论文中提及的全部车辆调度问题（Vehicle Scheduling Problem, VSP）分析数据集。分析涵盖多目的地案例研究，以及“高拥堵”与“低拥堵”两类示例的30趟行程。分析命名规则与博士论文一致。每个文件夹包含以下文件：
Aik.txt：本文件为一张矩阵，记录了每日每个时段（Tk=0:23）下，各目的地的单向行程时长（单位：小时）。
Cdik.txt：本文件为一张矩阵，记录了每日每个时段（Tk=0:23）下，各目的地的返程驾驶员时间成本（单位：英镑）。
Di.txt：本文件为一张列表，列明了各目的地的“最晚交付”限制要求。
Fc.txt：本文件为一张矩阵，记录了每日每个时段（Tk=0:23）下，各目的地的返程燃油消耗量（单位：升）。
Pik.txt：本文件为一张矩阵，记录了每日每个时段（Tk=0:23）下，各目的地的返程行程时长（单位：小时）。
Ri.txt：本文件为一张列表，列明了各目的地的“最早到达”配送限制要求。
Tk.txt：本文件为一张列表，列明了数据所对应的每日时段（Tk=0:23）。
TDFSVSP.py：本文件为Python程序，使用Gurobi商业求解器运行优化分析。
Output.xlsx：本文件为分析的Excel输出结果，包含分析得出的最优解的汇总信息与输出矩阵。
第5章：集成化运营模式变革的机遇
本文件夹包含博士论文中提及的全部车辆调度问题（Vehicle Scheduling Problem, VSP）分析数据集。分析涵盖多目的地案例研究，以及“高拥堵”与“低拥堵”两类示例的30趟行程。分析命名规则与博士论文一致。其中：RB=基准速度策略；S1：速度策略1（最高时速80km/h）；S2：速度策略2（最高时速70km/h）；Free：将速度策略建模为决策变量。
除标注为“Free”的分析外，其余每个文件夹包含以下文件：Aik.txt、Cdik.txt、Di.txt、Fc.txt、Pik.txt、Ri.txt、Tk.txt、TDFSVSP.py、Output.xlsx，各文件含义与第4章中一致。
在标注为“Free”的文件夹中，上述Aik、Cdik、Fc与Pik文件的后缀分别为0、1或2，分别对应前述的基准策略、S1与S2速度策略。