K-Means Clustering of Road Segments Based on Traffic Volume and Pavement Distress Index

Project info

• Field of studyCivil Engineering and Transportation
• LevelUndergraduate thesis (S1)
• Method typeClustering
• Estimated build time± 3 working days
• Toolsscikit-learn KMeans, silhouette_score, StandardScaler, make_blobs
• PriceUS$19 - US$116

Research goal

Group road segments based on traffic volume and distress index with K-Means Clustering to uncover group patterns.

Variables

Example result

• 3Clusters k
• 0.768Silhouette
• 0.3Davies-Bouldin
• 948.536Calinski-Harabasz
• 67.746Inertia (std.)
• 0.478ARI vs. true groups

K-Means forms 3 clusters on Road segments, Traffic volume and Distress index with a silhouette of 0.768.

Computed from illustrative synthetic data, not from a real study.

About K-Means Clustering

K-Means partitions data into k clusters by minimising the distance of points to their cluster centres. It suits segmentation of numeric data, such as customers or regions, when the number of clusters can be estimated.

Steps

1. 1Select numeric features and standardise them
2. 2Run K-Means for several values of k
3. 3Choose k with the elbow method (inertia) and silhouette score
4. 4Fit the final model with several initialisations
5. 5Profile each cluster (feature means)
6. 6Plot the clusters and their centres

Assumptions to check

• Clusters are roughly spherical and similar in size
• Features are standardised because it uses Euclidean distance
• Outliers are handled since they affect cluster centres

Reading and reporting the results

Report the reason for choosing k (elbow and silhouette) and describe each cluster in substantive terms. Remind that clusters are exploratory findings, not absolute truth.

Example Python code

A short starting script with this project's variable names. The delivered version is adapted to your data and fully commented.

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import make_blobs
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import silhouette_score

X, _ = make_blobs(n_samples=400, centers=4, cluster_std=1.0, random_state=15)
X = StandardScaler().fit_transform(X)
traffic_volume = X[:, 0]
distress_index = X[:, 1]

ks = range(2, 9)
inertia, sil = [], []
for k in ks:
    km = KMeans(n_clusters=k, n_init=10, random_state=15).fit(X)
    inertia.append(km.inertia_)
    sil.append(silhouette_score(X, km.labels_))
    print("k =", k, "inertia =", round(km.inertia_, 1), "silhouette =", round(sil[-1], 3))

best_k = list(ks)[int(np.argmax(sil))]
road_segments = KMeans(n_clusters=best_k, n_init=10, random_state=15).fit(X)
print("Chosen k:", best_k)

fig, ax = plt.subplots(1, 2, figsize=(10, 4))
ax[0].plot(list(ks), inertia, marker="o"); ax[0].set_xlabel("k"); ax[0].set_ylabel("Inertia")
ax[1].scatter(traffic_volume, distress_index, c=road_segments.labels_, s=15)
ax[1].scatter(*road_segments.cluster_centers_.T, c="r", marker="x", s=80)
plt.tight_layout()
plt.show()

Packages and prices

Choose how much help you need. Each higher package includes everything in the lower ones. Prices exclude 11% VAT.

• Payment: 50% down payment before work starts, balance after the deliverables are handed over
• Tax: Prices exclude 11% VAT
• Data and academic integrity: Results are computed from your own research data; you write and stand behind your manuscript, we help with the analysis and your understanding of it
• Revisions: Up to 2 revisions are included in the price (Package 3: 3 rounds)

Frequently asked questions