muthuk1 commited on
Commit
0c0a658
·
verified ·
1 Parent(s): 7327b7b

Add model card for Workload Scoring Model

Browse files
Files changed (1) hide show
  1. README.md +85 -0
README.md ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: sklearn
3
+ tags:
4
+ - logistics
5
+ - xgboost
6
+ - route-optimization
7
+ - sklearn
8
+ - tabular-regression
9
+ - workload
10
+ - fairrelay
11
+ datasets:
12
+ - Cainiao-AI/LaDe-D
13
+ - electricsheepafrica/africa-synth-retail-and-ecommerce-last-mile-delivery-data-nigeria
14
+ license: mit
15
+ ---
16
+
17
+ # FairRelay — Workload Scoring Model
18
+
19
+ Part of the **[FairRelay](https://github.com/MUTHUKUMARAN-K-1/FairRelay)** AI logistics platform.
20
+
21
+ ## Model Description
22
+
23
+ Workload Scoring Model
24
+
25
+ **Type**: XGBRegressor Pipeline
26
+ **Framework**: scikit-learn Pipeline + XGBoost
27
+ **Task**: Regression
28
+
29
+ ## Performance
30
+
31
+ - **R²**: 0.9969
32
+ - **MAE**: 4.2338
33
+ - **RMSE**: 5.3418
34
+ - **CV R² (5-fold)**: 0.9969 ± 0.0001
35
+
36
+
37
+ ## Input Features
38
+
39
+ | Feature | Importance |
40
+ |---------|-----------|
41
+ | `num_packages` | 0.0205 |
42
+ | `total_weight_kg` | 0.0118 |
43
+ | `num_stops` | 0.3539 |
44
+ | `avg_fragility` | 0.0035 |
45
+ | `total_distance_km` | 0.0008 |
46
+ | `route_difficulty_score` | 0.5454 |
47
+ | `estimated_time_minutes` | 0.0640 |
48
+
49
+
50
+ ## Usage
51
+
52
+ ```python
53
+ from skops import io as sio
54
+ from huggingface_hub import hf_hub_download
55
+
56
+ # Download and load
57
+ model_path = hf_hub_download(repo_id="muthuk1/fairrelay-workload-scoring", filename="model.skops")
58
+ untrusted = sio.get_untrusted_types(file=model_path)
59
+ model = sio.load(model_path, trusted=untrusted)
60
+
61
+ # Predict
62
+ import numpy as np
63
+ features = np.array([[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]])
64
+ prediction = model.predict(features)
65
+ ```
66
+
67
+ ## Training Data
68
+
69
+ - **Cainiao-AI/LaDe-D**: Real last-mile delivery data from Shanghai (KDD 2023)
70
+ - **Africa Synth Last-Mile**: Synthetic Nigerian delivery data
71
+ - **FairRelay Synthetic**: Physics-informed synthetic data calibrated to FairRelay's deterministic formulas
72
+
73
+ ## Part of FairRelay
74
+
75
+ FairRelay is an AI-powered logistics platform for fair load consolidation and dispatch:
76
+ - 🚚 5-agent load consolidation pipeline (KMeans + OR-Tools CP-SAT)
77
+ - ⚖️ 8-agent fair dispatch pipeline (Gini optimization)
78
+ - 📊 XGBoost effort prediction + Thompson Sampling bandit
79
+ - 🌱 EV-aware routing with battery constraints
80
+
81
+ Built for **LogisticsNow Hackathon 2026** — Challenge #5: AI Load Consolidation
82
+
83
+ ## License
84
+
85
+ MIT