In this study, we employ hiking function to generate a local travel network connecting each settlement in the landscape with a specified number of its nearest neighbours. In order to determine the traffic value for each path in the network, a kernel density estimate was employed to handle the potential overlap of roads or proximity of multiple paths. 在本研究中，我们采用徒步函数生成一个本地旅行网络，将景观中的每个定居点与其指定数量的最近邻居连接起来。为了确定网络中每条路径的交通值，我们使用了核密度估计来处理道路的潜在重叠或多条路径的接近性。The paths were transformed into a series of evenly spaced points (100 m intervals) with each point being assigned the same traffic value and then the resulting traffic distribution within the study area was visualized through a weighted kernel density estimate. 将路径转换成一系列间隔均匀的点（间隔 100 米），每个点分配相同的交通值，然后通过加权核密度估计将研究区域内的交通分布可视化。
使用forts 碉堡 作为节点：
As the path of the road may exhibit a strong correlation with the forts it linked, the identification of nodes and construction of a spatial model of the road can aid in verifying the existence of road remains, which may be challenging to discover through fieldwork. 由于道路的路径可能与其所连接的堡垒表现出很强的相关性，因此识别节点和构建道路的空间模型可以帮助验证道路遗迹的存在，而这可能很难通过实地考察来发现。（之所以不直接考察道路，而考察堡壘，是因為很多泥土的道路現在已經不存在了，但是堡壘卻還在。）
时间跨度：
In terms of chronology, our study mainly focuses on the Western Han period (202 BCE – _8 CE). Although the Zhidao was originally constructed during the Qin Dynasty (221–207 BCE), achieving a more detailed chronological division of forts into distinct periods such as the Qin, Western Han, and Eastern Han (25–220 CE) proves challenging due to the coarse resolution of the archaeological evidence. 在年代学方面，我们的研究主要集中在西汉时期（公元前 202 年 - 公元 8 年）。虽然直道最初建于秦朝（公元前 221-207 年），但由于考古证据的分辨率较差，要将堡垒更详细地划分为秦、西汉和东汉（公元 25-220 年）等不同时期，是一项艰巨的任务。——囿於幾個原因，將秦漢的ports直接看作西漢的ports。