📌 Tentang Sesi Ini: Transportasi adalah aktivitas logistik dengan biaya terbesar (40-60% dari total biaya logistik). Sesi ini membahas infrastruktur transportasi Indonesia, operasional transportasi, dan bagaimana AI merevolusi route optimization. Kita akan belajar dari teori (Bowersox Ch 8-9, Chopra Ch 14) hingga praktik dengan simulasi AI.
🚚 PART A: Transportation Fundamentals
Memahami peran strategis transportasi dalam supply chain: moda, infrastruktur, dan biaya
1. Peran Strategis Transportasi dalam SCM (Bowersox Ch 8)
▼
2. Moda Transportasi: Karakteristik & Trade-off (Bowersox Ch 9)
▼
Pemilihan moda transportasi adalah keputusan strategis yang mempengaruhi cost, speed, reliability, dan capability.
🎯 5 Moda Transportasi Utama:
🎯 5 Moda Transportasi: Perbandingan Komprehensif
SHIP (Laut)
Speed: ⭐☆☆☆☆
Cost: ⭐⭐⭐⭐⭐
Capacity: ⭐⭐⭐⭐⭐
Reliability: ⭐⭐⭐☆☆
Best for: Bulk cargo, international trade, antar-pulau
RAIL (Kereta)
Speed: ⭐⭐⭐☆☆
Cost: ⭐⭐⭐⭐☆
Capacity: ⭐⭐⭐⭐☆
Reliability: ⭐⭐⭐⭐☆
Best for: Bulk commodities, jarak jauh darat
TRUCK (Darat)
Speed: ⭐⭐⭐⭐☆
Cost: ⭐⭐⭐☆☆
Capacity: ⭐⭐⭐☆☆
Reliability: ⭐⭐⭐⭐☆
Best for: Door-to-door, regional distribution
AIR (Udara)
Speed: ⭐⭐⭐⭐⭐
Cost: ⭐☆☆☆☆
Capacity: ⭐⭐☆☆☆
Reliability: ⭐⭐⭐⭐⭐
Best for: High-value, urgent, international
MOTOR (Last-Mile)
Speed: ⭐⭐⭐⭐⭐
Cost: ⭐⭐⭐⭐☆
Capacity: ⭐☆☆☆☆
Reliability: ⭐⭐⭐⭐☆
Best for: E-commerce, food delivery, dokumen
📊 Perbandingan Detail Moda Transportasi:
| Karakteristik | 🚢 Ship | 🚂 Rail | 🚛 Truck | ✈️ Air | 🛵 Motor |
|---|---|---|---|---|---|
| Speed (km/hari) | 200-400 | 400-600 | 400-800 | 5,000-10,000 | 100-200 |
| Cost (Rp/kg/km) | 50-150 | 100-250 | 200-500 | 2,000-5,000 | 500-1,500 |
| Capacity (ton) | 10,000-500,000 | 1,000-5,000 | 5-30 | 10-100 | 0.05-0.2 |
| Reliability (%) | 75-85 | 85-90 | 90-95 | 95-98 | 90-95 |
| Accessibility | Port only | Station only | Door-to-door | Airport only | Door-to-door |
| Best For | Bulk, international | Bulk, long distance | Regional, flexible | Urgent, high-value | Small, urban |
🎯 Decision Framework: Pilih Moda yang Tepat
| Faktor Keputusan | Pertimbangan | Rekomendasi Moda | Contoh Kasus |
|---|---|---|---|
| Product Value | Tinggi (> Rp 1 juta/kg) | Air freight | Elektronik, obat-obatan, perhiasan |
| Product Value | Rendah (< Rp 100rb/kg) | Ship atau Rail | Beras, semen, batubara |
| Urgency | Sangat urgent (< 24 jam) | Air freight | Spare part kritis, dokumen |
| Urgency | Normal (3-7 hari) | Truck atau Rail | FMCG, retail goods |
| Distance | Antar-pulau/internasional | Ship atau Air | Jawa-Sumatera, Indonesia-China |
| Distance | Dalam pulau (< 500 km) | Truck | Jakarta-Bandung, Surabaya-Malang |
| Volume | Bulk (> 100 ton) | Ship atau Rail | Komoditas, bahan baku |
| Volume | Small (< 1 ton) | Truck atau Motor | E-commerce, parcel |
🇮🇩 Contoh Indonesia: Modal Split di Indonesia
Distribusi Moda Transportasi Barang (2024):
- Truck (Road): 70-75% – Dominan untuk distribusi domestik
- Ship (Sea): 15-20% – Untuk antar-pulau dan ekspor-impor
- Rail: 3-5% – Terbatas di Jawa (batubara, komoditas)
- Air: 1-2% – Untuk barang bernilai tinggi & urgent
- Motor (Last-Mile): Growing rapidly – E-commerce boom
Tantangan Modal Split Indonesia:
- Over-reliance on truck: 70%+ barang diangkut truk → kemacetan, emisi tinggi
- Under-utilized rail: Hanya 3-5% vs Thailand 20%+ → perlu investasi infrastruktur
- Coastal shipping: Potensi besar tapi belum optimal → perlu insentif
Best Practice: Multimodal Transportation
- Konsep: Kombinasi 2+ moda dalam satu shipment untuk optimasi cost & speed
- Contoh: Shanghai → Tanjung Priok (ship) → Jakarta DC (truck) → Customer (motor)
- Benefit: Cost -30%, lead time optimal, flexibility tinggi
💡 Insight: Tidak ada “best mode” universal. Pemilihan moda harus berdasarkan trade-off analysis antara cost, speed, reliability, dan capability. Multimodal transportation seringkali menjadi solusi optimal untuk Indonesia yang kompleks.
3. Infrastruktur Transportasi Indonesia: Kondisi & Tantangan
▼
🏗️ Kondisi Infrastruktur Transportasi Indonesia (2024):
| Jenis Infrastruktur | Kondisi Saat Ini | Tantangan | Government Plan |
|---|---|---|---|
| 🛣️ Jalan Tol | • 2,000+ km (2024) • Terkonsentrasi di Jawa (70%) • Masih banyak ruas belum terhubung |
• Disparitas Jawa vs luar Jawa • Biaya tol tinggi • Konektivitas ke pelabuhan |
• Target 3,500 km (2029) • Trans-Sumatera, Trans-Kalimantan • Tol laut integration |
| 🚂 Rel Kereta | • 8,000+ km (mostly Java) • Kereta api barang terbatas • Kecepatan rendah (40-60 km/jam) |
• Infrastruktur tua • Single track di banyak ruas • Kurang investasi |
• Double track Java • Kereta cepat Jakarta-Bandung • Revitalisasi jalur barang |
| 🚢 Pelabuhan | • 1,200+ pelabuhan • Tanjung Priok (Jakarta) = 30% volume • Banyak pelabuhan kecil |
• Dwelling time masih tinggi • Kapasitas terbatas • Alat bongkar muat tua |
• Pengembangan hub port • Digitalisasi (INSW) • Tol laut program |
| ✈️ Bandara | • 270+ bandara • Soetta (Jakarta) = 40% traffic • Banyak bandara perintis |
• Overcapacity di bandara utama • Konektivitas timur Indonesia • Cargo capacity terbatas |
• Expansion Soetta, Juanda • Bandara baru (DIY, etc.) • Cargo hub development |
| 🚛 Logistik Center | • Terkonsentrasi di Jabodetabek • Warehouse modern growing • 3PL industry berkembang |
• Disparitas regional • Teknologi adopsi lambat • Skill gap |
• Pengembangan KLBI • Digital logistics platform • Training & certification |
📊 World Bank Logistics Performance Index (LPI) 2023:
| Negara | Rank | Score | Key Strength | Key Weakness |
|---|---|---|---|---|
| 🇩🇪 Germany | 1 | 4.1 | Excellent infrastructure | – |
| 🇸🇬 Singapore | 2 | 4.0 | World-class port | – |
| 🇯🇵 Japan | 3 | 3.9 | Reliability & efficiency | – |
| … | … | … | … | … |
| 🇲🇾 Malaysia | 36 | 3.5 | Port infrastructure | Customs efficiency |
| 🇹🇭 Thailand | 34 | 3.5 | Manufacturing hub | Infrastructure gap |
| 🇻🇳 Vietnam | 43 | 3.3 | Competitive cost | Infrastructure quality |
| 🇮🇩 Indonesia | 61 | 3.1 | International shipments | Infrastructure & customs |
| 🇵🇭 Philippines | 65 | 3.0 | Growing economy | Archipelago challenges |
🇮🇩 Studi Kasus: Tol Laut Program
Latar Belakang: Disparitas harga antara Jawa dan luar Indonesia mencapai 20-30% karena biaya logistik tinggi.
Program Tol Laut (sejak 2015):
- Konsep: Subsidi rute pelayaran reguler dari Jawa ke luar Jawa
- Routes: 26 rute (2024) menghubungkan Jawa-Sumatera-Kalimantan-Sulawesi-Papua
- Subsidi: Pemerintah subsidi biaya operasional kapal
- Target: Menurunkan disparitas harga menjadi < 10%
Hasil (2015-2024):
- Disparitas harga turun dari 25% → 15% (masih di atas target)
- Volume barang via tol laut: 500,000+ TEU/tahun
- Lead time: 7-10 hari → 4-6 hari
- Biaya logistik: -20% untuk rute tol laut
Tantangan:
- Subsidi berkelanjutan (beban APBN)
- Konektivitas darat dari pelabuhan ke customer
- Kapasitas pelabuhan di daerah terbatas
- Koordinasi antar-stakeholder
⚠️ Realita Transportasi Indonesia:
- Empty Return Trips: 40-50% truk kembali kosong → biaya per trip naik 2x
- Kemacetan: Jakarta = 40-60 km/jam average speed → delay & biaya tinggi
- Pajak Daerah: 10+ pungutan liar per trip → biaya tak terduga
- Overloading: Banyak truk overload → kerusakan jalan, safety risk
- Solusi: Digital platform (seperti Kargo.tech, Waresix) untuk match supply-demand
4. Transportation Optimization: Traditional Approaches
▼
🎯 4 Masalah Optimasi Transportasi Klasik:
| Masalah | Deskripsi | Metode Solusi | Contoh Aplikasi |
|---|---|---|---|
| Transportation Problem | Minimalkan biaya pengiriman dari multiple suppliers ke multiple destinations | Linear programming, Northwest Corner, Vogel’s Approximation | Alokasi barang dari 5 pabrik ke 20 distributor |
| Vehicle Routing Problem (VRP) | Tentukan rute optimal untuk fleet kendaraan yang melayani multiple customers | Heuristics, metaheuristics, exact algorithms | 10 truk melayani 100 customer di Jakarta |
| Traveling Salesman Problem (TSP) | Temukan rute terpendek untuk mengunjungi n locations dan kembali ke start | Nearest neighbor, genetic algorithm, simulated annealing | Salesman mengunjungi 20 kota di Jawa |
| Network Design Problem | Tentukan lokasi fasilitas (warehouse, DC) dan aliran barang optimal | Mixed-integer programming, heuristic | Dimana buka 3 warehouse baru untuk serve Indonesia |
📊 Traditional Optimization Methods:
| Method | Kelebihan | Kekurangan | Kapan Digunakan |
|---|---|---|---|
| Linear Programming | • Optimal solution guaranteed • Well-understood mathematically • Many software available |
• Requires linear relationships • Computationally expensive for large problems • Static (no real-time) |
Small-medium problems, stable environment |
| Heuristics | • Fast computation • Can handle large problems • Flexible |
• No guarantee of optimality • Quality depends on heuristic design • May get stuck in local optimum |
Large problems, time-constrained |
| Simulation | • Can model complex systems • What-if analysis • Visual output |
• Computationally expensive • Requires detailed data • Doesn’t provide optimal solution |
Complex systems, scenario analysis |
🇮🇩 Contoh Indonesia: Traditional Route Planning di JNE
Konteks: JNE memiliki 500+ truk untuk distribusi paket di Jawa.
Traditional Approach (Pre-2018):
- Manual planning: Dispatcher buat rute berdasarkan pengalaman
- Static routes: Rute tetap, tidak berubah harian
- Zone-based: Jakarta dibagi 5 zona, setiap truk handle 1 zona
- Tools: Excel + Google Maps manual
Hasil:
- Average distance per truk: 150 km/hari
- Fuel consumption: 15 liter/100km
- Delivery time: 2-3 hari untuk Jawa
- Empty return trips: 35%
- Driver overtime: 20 jam/minggu
Limitasi:
- Tidak adaptif terhadap traffic real-time
- Tidak optimasi multi-stop
- Tidak consider time windows
- Human error dalam planning
💡 Insight: Traditional methods masih relevan untuk small-scale problems, tapi untuk operasi besar (100+ vehicles, 1000+ stops), kita butuh AI-powered optimization yang bisa handle complexity dan real-time changes.
⚙️ PART B: Transportation Operations
Operasional transportasi: route planning, fleet management, dan last-mile delivery
5. Route Planning & Scheduling (Chopra Ch 14)
▼
6. Fleet Management: Optimasi Armada
▼
🎯 5 Komponen Fleet Management:
| Komponen | Aktivitas | KPI | Tools |
|---|---|---|---|
| ① Vehicle Acquisition | • Buy vs lease decision • Vehicle specification • Fleet size determination |
• Total cost of ownership • Utilization rate |
• Financial modeling • Fleet optimization software |
| ② Maintenance | • Preventive maintenance • Breakdown repair • Spare parts inventory |
• Downtime % • Maintenance cost/vehicle • Mean time between failures |
• CMMS (Computerized Maintenance Management) • IoT sensors |
| ③ Fuel Management | • Fuel consumption monitoring • Fuel card management • Fuel efficiency programs |
• Liter/100km • Fuel cost/km • Idling time % |
• Fuel management systems • GPS tracking • Telematics |
| ④ Driver Management | • Recruitment & training • Performance monitoring • Safety programs |
• Driver turnover rate • Accident rate • Productivity (deliveries/driver) |
• HR systems • Driver scorecards • Dashcam |
| ⑤ Compliance | • Regulatory compliance • Insurance management • Documentation |
• Violations • Insurance claims • Audit findings |
• Compliance software • Digital documentation |
💰 Total Cost of Ownership (TCO) per Vehicle:
| Cost Component | % dari TCO | Contoh (Truck CDD) | Strategi Pengendalian |
|---|---|---|---|
| Depresiasi | 20-25% | Rp 50-75 juta/tahun | Optimal replacement cycle (5-7 tahun) |
| Bahan Bakar | 25-35% | Rp 75-100 juta/tahun | Fuel-efficient driving, route optimization |
| Driver Salary | 20-25% | Rp 60-75 juta/tahun | Performance-based incentives |
| Maintenance | 10-15% | Rp 30-45 juta/tahun | Preventive maintenance program |
| Insurance & Tax | 5-10% | Rp 15-30 juta/tahun | Fleet discount, safe driving record |
| Toll & Parking | 3-5% | Rp 10-15 juta/tahun | Route optimization, RFID toll cards |
| TOTAL TCO | 100% | Rp 240-340 juta/tahun | Target: Rp 8,000-12,000/km |
🇮🇩 Contoh Indonesia: Fleet Management di Blue Bird
Konteks: Blue Bird mengelola 25,000+ taksi di Indonesia.
Fleet Management System:
- GPS Tracking: Real-time monitoring semua kendaraan
- Telematics: Monitor speed, braking, idling, fuel consumption
- Maintenance scheduling: Automated alerts untuk service
- Driver scorecard: Performance monitoring (safety, efficiency, customer rating)
Hasil:
- Fuel consumption: -15% (eco-driving program)
- Accident rate: -40% (safety training + monitoring)
- Maintenance cost: -20% (preventive vs reactive)
- Vehicle utilization: 85% (optimal scheduling)
- Driver retention: 85% (performance incentives)
7. Last-Mile Delivery: Tantangan & Solusi
▼
Definisi: Last-mile delivery adalah tahap terakhir dalam pengiriman barang dari facility ke customer akhir. Ini adalah tahap paling mahal (50-60% dari total transportation cost) dan paling kompleks karena volume kecil, banyak stops, dan customer expectations tinggi.
⚠️ 5 Tantangan Last-Mile Delivery:
| Tantangan | Deskripsi | Impact |
|---|---|---|
| ① High Cost | 50-60% dari total transportation cost, bahkan untuk jarak pendek | • Margin tipis atau rugi • Customer tidak mau bayar full cost • Subsidi silang dari margin produk |
| ② Complexity | Banyak stops, small parcels, time windows, access issues | • Route planning sulit • Low vehicle utilization • High failure rate (failed deliveries) |
| ③ Customer Expectations | Same-day/next-day delivery, real-time tracking, flexible delivery | • Pressure pada operations • Need for technology investment • Service level agreements ketat |
| ④ Urban Congestion | Kemacetan, parking terbatas, restricted zones | • Delay & unpredictability • Higher fuel consumption • Driver frustration |
| ⑤ Failed Deliveries | Customer tidak ada di rumah, alamat salah, refused delivery | • 10-20% first-attempt failure rate • Re-delivery cost 2-3x • Customer dissatisfaction |
🎯 6 Solusi Last-Mile Innovation:
| Solusi | Deskripsi | Keuntungan | Contoh Indonesia |
|---|---|---|---|
| ① Crowdshipping | Gunakan gig workers (ojol) untuk delivery | • Flexible capacity • Low fixed cost • Fast delivery |
Gojek, Grab, Lalamove |
| ② Pickup Points | Customer ambil di lokasi convenience (minimarket, locker) | • Reduce failed deliveries • Lower cost per delivery • 24/7 availability |
Indomaret, Alfamart, PopBox |
| ③ Micro-Fulfillment | Small warehouses di dalam kota untuk faster delivery | • Same-day delivery • Lower last-mile distance • Better inventory turnover |
Aston, Fulfillment.id |
| ④ Route Optimization AI | AI untuk optimize routes real-time | • Distance -20-30% • Time -15-25% • Fuel -15-20% |
J&T, SiCepat, Anteraja |
| ⑤ Electric Vehicles | Motor/mobil listrik untuk delivery | • Lower operating cost • Zero emission • Access to restricted zones |
Volta (Gojek), GrabElectric |
| ⑥ Drone/Robot | Autonomous delivery (masih pilot) | • Ultra-fast delivery • No labor cost • 24/7 operation |
Belum ada di Indonesia (regulasi) |
🇮🇩 Contoh Indonesia: Last-Mile Innovation di Tokopedia
Konteks: Tokopedia melakukan 5+ juta deliveries/hari di seluruh Indonesia.
Last-Mile Strategy:
- Multi-courier integration: 20+ courier partners (JNE, J&T, SiCepat, Pos, dll)
- AI-powered allocation: Order di-assign ke courier optimal berdasarkan lokasi, cost, SLA
- Pickup points: 50,000+ titik (Indomaret, Alfamart, mitra)
- Real-time tracking: Customer bisa track order real-time
- Flexible delivery: Customer bisa pilih waktu & lokasi delivery
Hasil:
- Average delivery time: 2.5 hari (Jawa), 4-5 hari (luar Jawa)
- On-time delivery rate: 94%
- Failed delivery rate: 5% (vs industry avg 15%)
- Customer satisfaction: 4.6/5
- Cost per delivery: Rp 12,000-18,000
Innovation:
- Tokopedia Instant: Same-day delivery via Gojek/Grab (1-2 jam)
- Tokopedia Now: Next-day delivery untuk produk tertentu
- Tokopedia Warehouse: Fulfillment by Tokopedia untuk merchant
💡 Insight: Last-mile delivery adalah battlefield utama e-commerce. Perusahaan yang bisa solve last-mile challenge (fast, cheap, reliable) akan menang. Solusi seringkali hybrid: kombinasi own fleet + 3PL + crowdshipping + pickup points.
8. Cross-Docking & Hub-and-Spoke Networks
▼
🔄 Cross-Docking: Konsep & Manfaat
| Aspek | Deskripsi | Benefit |
|---|---|---|
| Definisi | Proses memindahkan barang dari inbound ke outbound vehicle tanpa penyimpanan di warehouse | • Eliminate storage cost • Reduce handling • Faster throughput |
| Flow | Inbound → Unload → Sort → Consolidate → Load → Outbound (dalam hitungan jam) | • Lead time reduction • Inventory reduction • Space utilization |
| Requirements | • Advanced IT systems (WMS, TMS) • Reliable suppliers • Accurate demand forecasting • Fast transportation |
• Lower total logistics cost • Better responsiveness • Reduced obsolescence |
🕸️ Hub-and-Spoke Network Design:
🕸️ Hub-and-Spoke Network: Efisiensi melalui Konsolidasi
🏭 HUB (Consolidation Center)
Jakarta DC / Surabaya Hub
🏪
Spoke 1
Bandung
🏪
Spoke 2
Semarang
🏪
Spoke 3
Yogyakarta
🏪
Spoke 4
Surabaya
✅ Keuntungan Hub-and-Spoke
• Konsolidasi volume → economies of scale
• Reduced transportation cost (-20-30%)
• Better service level (faster delivery)
• Simplified network management
🇮🇩 Contoh Indonesia: Hub-and-Spoke di JNE
Konteks: JNE mengoperasikan 5,000+ pickup points di seluruh Indonesia.
Network Structure:
- Main Hub: Jakarta (Cengkareng) – sorting center utama
- Regional Hubs: 15 hub di kota besar (Surabaya, Medan, Makassar, dll)
- Sub-Hubs: 50+ sub-hub di kota menengah
- Service Points: 5,000+ agen & pickup points
Flow:
- Customer drop-off di service point
- Service point → Sub-hub (harian)
- Sub-hub → Regional hub (harian)
- Regional hub → Main hub Jakarta (harian, via udara/laut)
- Main hub → Sorting & routing
- Main hub → Destination regional hub
- Regional hub → Sub-hub destination
- Sub-hub → Service point destination
- Service point → Customer (delivery)
Hasil:
- Coverage: 95% populasi Indonesia
- Lead time: 1-2 hari (Jawa), 3-5 hari (luar Jawa)
- Volume: 1+ juta paket/hari
- Cost efficiency: 30% lebih rendah vs point-to-point
🤖 PART C: AI-Powered Route Optimization
Revolusi transportasi dengan AI: dari traditional methods ke intelligent optimization
9. Traditional vs AI Route Optimization (Chopra Ch 14)
▼
10. Workshop: Simulasi Rute Transportasi dengan AI
▼
Tugas Mahasiswa: Anda adalah logistics manager di PT FMCG Indonesia yang harus mendistribusikan produk dari warehouse di Jakarta ke 15 retailer di Jabodetabek.
📋 Problem Statement:
Data:
- Warehouse: Jakarta Utara (origin)
- Customers: 15 retailer di Jabodetabek
- Vehicle: 2 truk CDD (capacity 2 ton each)
- Constraints:
- Vehicle capacity: 2 ton per truck
- Driver working hours: max 10 jam/hari
- Time windows: Customer hanya bisa receive 08:00-17:00
- Service time: 15 menit per customer
- Objective: Minimize total distance & time, serve all customers within constraints
Customer Data:
| ID | Location | Distance from Warehouse (km) | Demand (ton) | Time Window |
|---|---|---|---|---|
| C1 | Jakarta Pusat | 10 | 0.3 | 08:00-12:00 |
| C2 | Jakarta Selatan | 15 | 0.4 | 09:00-13:00 |
| C3 | Jakarta Barat | 12 | 0.2 | 08:00-11:00 |
| C4 | Jakarta Timur | 18 | 0.5 | 10:00-14:00 |
| C5 | Bekasi | 25 | 0.6 | 09:00-15:00 |
| C6 | Depok | 22 | 0.4 | 10:00-16:00 |
| C7 | Tangerang | 20 | 0.5 | 08:00-14:00 |
| C8 | Bogor | 35 | 0.7 | 11:00-17:00 |
| C9 | Jakarta Pusat (2) | 11 | 0.3 | 13:00-17:00 |
| C10 | Bekasi (2) | 28 | 0.4 | 13:00-17:00 |
| C11 | Depok (2) | 24 | 0.3 | 08:00-12:00 |
| C12 | Tangerang (2) | 22 | 0.4 | 14:00-17:00 |
| C13 | Jakarta Selatan (2) | 16 | 0.2 | 14:00-17:00 |
| C14 | Jakarta Barat (2) | 13 | 0.3 | 15:00-17:00 |
| C15 | Bogor (2) | 38 | 0.5 | 08:00-13:00 |
🎯 Langkah 1: Manual Route Planning (15 menit)
📝 Tugas:
1. Bagi 15 customers ke 2 truk (consider capacity & time windows)
2. Tentukan urutan kunjungan untuk setiap truk
3. Hitung total distance dan time
4. Pastikan semua constraints terpenuhi
Hint: Mulai dengan customers yang punya time window ketat (pagi), lalu cluster by geography
1. Bagi 15 customers ke 2 truk (consider capacity & time windows)
2. Tentukan urutan kunjungan untuk setiap truk
3. Hitung total distance dan time
4. Pastikan semua constraints terpenuhi
Hint: Mulai dengan customers yang punya time window ketat (pagi), lalu cluster by geography
🎯 Langkah 2: AI-Powered Optimization dengan ChatGPT (10 menit)
📝 Prompt untuk AI:
“Saya punya vehicle routing problem dengan data berikut:
Origin: Warehouse di Jakarta Utara
Customers: 15 retailers dengan data lokasi, demand, dan time window (lihat tabel di atas)
Vehicles: 2 truk CDD, capacity 2 ton each
Constraints:
– Vehicle capacity: 2 ton
– Driver working hours: max 10 jam
– Service time: 15 menit per customer
– Average speed: 30 km/jam (Jakarta traffic)
Objective: Minimize total distance
Tolong:
1. Assign customers ke 2 truk (consider capacity & time windows)
2. Optimize route sequence untuk setiap truk
3. Hitung total distance, time, dan utilization
4. Berikan timeline untuk setiap route (arrival & departure time per customer)
5. Identifikasi potential issues (time window violations, capacity issues)
Gunakan algoritma yang sesuai (nearest neighbor, savings algorithm, atau metaheuristic).”
“Saya punya vehicle routing problem dengan data berikut:
Origin: Warehouse di Jakarta Utara
Customers: 15 retailers dengan data lokasi, demand, dan time window (lihat tabel di atas)
Vehicles: 2 truk CDD, capacity 2 ton each
Constraints:
– Vehicle capacity: 2 ton
– Driver working hours: max 10 jam
– Service time: 15 menit per customer
– Average speed: 30 km/jam (Jakarta traffic)
Objective: Minimize total distance
Tolong:
1. Assign customers ke 2 truk (consider capacity & time windows)
2. Optimize route sequence untuk setiap truk
3. Hitung total distance, time, dan utilization
4. Berikan timeline untuk setiap route (arrival & departure time per customer)
5. Identifikasi potential issues (time window violations, capacity issues)
Gunakan algoritma yang sesuai (nearest neighbor, savings algorithm, atau metaheuristic).”
🎯 Langkah 3: Compare Manual vs AI Solution (10 menit)
📝 Pertanyaan untuk Diskusi:
1. Berapa % improvement AI solution vs manual solution?
2. Apa yang membuat AI solution lebih baik?
3. Apakah ada constraints yang dilanggar oleh AI solution?
4. Bagaimana jika ada disruption (traffic jam, customer cancel)?
5. Apa limitasi dari AI solution ini?
1. Berapa % improvement AI solution vs manual solution?
2. Apa yang membuat AI solution lebih baik?
3. Apakah ada constraints yang dilanggar oleh AI solution?
4. Bagaimana jika ada disruption (traffic jam, customer cancel)?
5. Apa limitasi dari AI solution ini?
🎯 Langkah 4: Advanced Scenario – Real-Time Disruption (10 menit)
📝 Scenario:
“Tiba-tiba ada disruption:
– C5 (Bekasi) cancel order
– Traffic jam di route ke C8 (Bogor) → travel time +50%
– C11 minta delivery lebih awal (time window berubah ke 07:00-10:00)
Tolong re-optimize routes dengan AI untuk handle disruptions ini. Berikan updated timeline dan calculate impact pada total cost.”
“Tiba-tiba ada disruption:
– C5 (Bekasi) cancel order
– Traffic jam di route ke C8 (Bogor) → travel time +50%
– C11 minta delivery lebih awal (time window berubah ke 07:00-10:00)
Tolong re-optimize routes dengan AI untuk handle disruptions ini. Berikan updated timeline dan calculate impact pada total cost.”
🇮🇩 Contoh Output: AI-Optimized Routes
Truck 1 Route:
- Warehouse → C3 (Jakarta Barat) → C7 (Tangerang) → C12 (Tangerang 2) → C14 (Jakarta Barat 2) → C1 (Jakarta Pusat) → C9 (Jakarta Pusat 2) → Return
- Total distance: 85 km
- Total time: 6.5 jam
- Capacity utilization: 1.9/2.0 ton (95%)
- All time windows met: ✅
Truck 2 Route:
- Warehouse → C11 (Depok 2) → C2 (Jakarta Selatan) → C6 (Depok) → C4 (Jakarta Timur) → C5 (Bekasi) → C10 (Bekasi 2) → C15 (Bogor) → C8 (Bogor 2) → C13 (Jakarta Selatan 2) → Return
- Total distance: 145 km
- Total time: 9.5 jam
- Capacity utilization: 2.0/2.0 ton (100%)
- All time windows met: ✅
Total Performance:
- Total distance: 230 km (vs manual 310 km → -26%)
- Total time: 16 jam (vs manual 20 jam → -20%)
- Fuel savings: 26% → Rp 500,000/hari
- Driver overtime: 0 jam (vs manual 4 jam)
⚠️ Limitasi AI Route Optimization:
- Data quality: AI hanya sebaik data yang dimasukkan. Data lokasi yang salah = rute yang salah
- Real-time data: AI tidak punya akses real-time traffic (butuh API seperti Google Maps)
- Human factors: AI tidak consider driver preference, local knowledge, customer relationship
- Over-reliance: Jangan blindly trust AI. Selalu verify dan use human judgment
- Solusi: Use AI sebagai decision support, bukan decision maker. Combine AI optimization dengan human expertise.
11. AI Tools & Implementation Guide
▼
🛠️ AI Tools for Route Optimization:
| Tool | Vendor | Features | Harga |
|---|---|---|---|
| Google Maps Platform | • Routes API • Distance Matrix API • Real-time traffic • Geocoding |
$5-10 per 1000 requests | |
| OptimoRoute | OptimoRoute | • Route optimization • Real-time tracking • Proof of delivery • Customer notifications |
$6-12/driver/month |
| Route4Me | Route4Me | • Multi-stop routing • GPS tracking • Route optimization • Mobile app |
$40-200/month |
| Onfleet | Onfleet | • Last-mile optimization • Real-time tracking • Customer experience • Analytics |
$150-500/month |
| Verizon Connect | Verizon | • Fleet management • Route optimization • GPS tracking • Compliance |
$25-45/vehicle/month |
| Custom AI Solution | In-house / Vendor | • Custom algorithms • Integration dengan existing systems • Full control • Scalable |
$50,000-500,000 (development) |
📊 Implementation Roadmap:
| Phase | Timeline | Activities | Deliverables |
|---|---|---|---|
| Phase 1: Assessment | 1-2 bulan | • Current state analysis • Data quality audit • Requirements gathering • ROI calculation |
• Assessment report • Business case • Requirements document |
| Phase 2: Selection | 1 bulan | • Vendor evaluation • Proof of concept • Pilot planning • Contract negotiation |
• Vendor shortlist • POC results • Selected vendor • Contract signed |
| Phase 3: Pilot | 2-3 bulan | • Pilot implementation • Data integration • User training • Performance monitoring |
• Pilot results • Lessons learned • Go/no-go decision • Rollout plan |
| Phase 4: Rollout | 3-6 bulan | • Full implementation • Data migration • Training all users • Change management |
• Live system • Trained users • Documentation • Support structure |
| Phase 5: Optimization | Ongoing | • Performance monitoring • Continuous improvement • Feature enhancements • Advanced analytics |
• KPI dashboards • Improvement reports • ROI tracking • Best practices |
🇮🇩 Contoh Indonesia: AI Implementation di SiCepat
Konteks: SiCepat melakukan 1+ juta deliveries/hari di Indonesia.
Implementation Journey:
- Phase 1 (2019): Assessment & vendor selection
- Phase 2 (2020): Pilot di Jakarta (50 drivers)
- Phase 3 (2021): Rollout ke Jawa (500 drivers)
- Phase 4 (2022): Full implementation nasional (5,000+ drivers)
- Phase 5 (2023+): Advanced features (predictive analytics, dynamic routing)
Technology Stack:
- Route optimization: Custom AI algorithm (genetic algorithm + local search)
- Real-time tracking: GPS + mobile app
- Integration: API dengan order management system
- Analytics: Dashboard untuk monitoring KPIs
Results:
- Distance per delivery: -28%
- Fuel consumption: -22%
- Delivery time: -35%
- Driver productivity: +40%
- Customer satisfaction: 4.8/5
- Cost savings: Rp 100 miliar/tahun
ROI:
- Total investment: Rp 25 miliar (3 tahun)
- Annual savings: Rp 100 miliar
- Payback period: 3 bulan
- ROI: 1200% dalam 3 tahun
💡 Best Practices for AI Implementation:
- Start small: Pilot dengan small fleet, learn, then scale
- Data first: Pastikan data quality sebelum implement AI
- User involvement: Libatkan drivers & dispatchers dari awal
- Change management: Training & communication adalah kunci
- Continuous improvement: AI bukan “set and forget”, perlu continuous tuning
- Measure ROI: Track KPIs dan calculate ROI secara regular
12. Future Trends: Transportation 2030
▼
🚀 7 Tren Masa Depan:
| Tren | Deskripsi | Timeline | Impact untuk Indonesia |
|---|---|---|---|
| 🚗 Autonomous Vehicles | Self-driving trucks & delivery vehicles | 2028-2035 | Masih 10-15 tahun. Regulasi & infrastruktur jadi tantangan utama. |
| 🚁 Drone Delivery | Autonomous drone untuk last-mile delivery | 2025-2030 | Potensi besar untuk area rural & kepulauan. Regulasi masih berkembang. |
| 🔋 Electric Vehicles | Electric trucks, vans, motorcycles | 2024-2030 | Sudah mulai. Government push untuk EV adoption. Charging infrastructure jadi kunci. |
| 🤖 AI & Machine Learning | Advanced route optimization, predictive analytics, autonomous decision-making | 2024-2028 | Sudah diadopsi early adopters. Akan jadi standard dalam 5 tahun. |
| 🌐 IoT & Connectivity | Connected vehicles, real-time tracking, smart infrastructure | 2024-2028 | Growing rapidly. 5G akan accelerate adoption. |
| 🌱 Sustainability | Green logistics, carbon-neutral transportation, circular economy | 2025-2035 | Mulai didorong regulasi & consumer pressure. ESG reporting jadi mandatory. |
| 🔗 Blockchain | Transparent supply chain, smart contracts, traceability | 2025-2030 | Early stage. Potensi untuk cross-border trade & compliance. |
💡 Prediksi untuk Indonesia 2030:
- EV adoption: 30% commercial vehicles akan jadi electric
- AI optimization: 80% logistics companies akan pakai AI untuk routing
- Autonomous: Masih early stage, tapi pilot projects di area terbatas
- Drone delivery: Commercial operation di area rural & kepulauan
- Sustainability: Carbon-neutral logistics jadi competitive advantage
- Cost reduction: Total logistics cost turun dari 24% → 15% GDP
- Job market: Traditional driver jobs turun 20%, tapi tech-enabled jobs naik 150%
📋 Ringkasan Eksekutif Sesi 06
- Peran Transportasi: Transportasi adalah aktivitas logistik dengan biaya terbesar (40-60% total cost). Menciptakan place & time utility, dan menjadi strategic lever untuk competitive advantage.
- 5 Moda Transportasi: Ship (slow, cheap, bulk), Rail (medium, bulk), Truck (flexible, door-to-door), Air (fast, expensive), Motor (last-mile, urban). Pemilihan berdasarkan trade-off cost, speed, reliability, capability.
- Infrastruktur Indonesia: Masih tertinggal (LPI rank 61). Tantangan: disparitas Jawa vs luar Jawa, empty return trips 40-50%, kemacetan, pajak daerah. Government initiatives: Tol Laut, infrastruktur, deregulasi.
- Transportation Optimization: 4 masalah klasik: Transportation Problem, VRP, TSP, Network Design. Traditional methods (LP, heuristics) masih relevan untuk small-scale, tapi AI diperlukan untuk large-scale.
- Route Planning: 4 tipe problems (point-to-point, TSP, VRP, VRPTW). Constraints: capacity, time windows, driver hours, vehicle availability, road restrictions. Objectives: minimize distance/time/cost, maximize service level.
- Fleet Management: 5 komponen: acquisition, maintenance, fuel, driver, compliance. TCO per truck CDD: Rp 240-340 juta/tahun. KPI: utilization, downtime, fuel efficiency, safety.
- Last-Mile Delivery: Tahap paling mahal (50-60% cost) dan kompleks. 5 tantangan: high cost, complexity, customer expectations, congestion, failed deliveries. 6 solusi: crowdshipping, pickup points, micro-fulfillment, AI optimization, EV, drone/robot.
- Cross-Docking & Hub-and-Spoke: Cross-docking eliminate storage, faster throughput. Hub-and-spoke network konsolidasi volume, reduce cost 20-30%. Contoh: JNE network dengan 15 regional hubs.
- AI Route Optimization: vs Traditional: real-time data, machine learning, scalable, adaptive, learning. 6 techniques: GA, ACO, SA, RL, Neural Networks, Hybrid. Contoh: J&T Express -25% distance, -20% fuel, ROI 500%.
- Workshop: Simulasi VRP dengan 15 customers, 2 trucks. Manual vs AI: AI -26% distance, -20% time. Advanced scenario: handle disruptions real-time.
- AI Tools: Google Maps, OptimoRoute, Route4Me, Onfleet, Verizon Connect, Custom solution. Implementation roadmap: 5 phases (Assessment → Selection → Pilot → Rollout → Optimization).
- Future Trends 2030: Autonomous vehicles, drone delivery, EV, AI/ML, IoT, sustainability, blockchain. Indonesia: 30% EV, 80% AI adoption, logistics cost 24% → 15% GDP.
📚 Referensi
- Bowersox, D.J., Closs, D.J., & Cooper, M.B. (2019). Supply Chain Logistics Management (5th ed.). McGraw-Hill. Chapter 8: Transportation, Chapter 9: Transportation Operations
- Chopra, S., & Meindl, P. (2023). Supply Chain Management: Strategy, Planning, and Operation (8th ed.). Pearson. Chapter 14: Transportation in a Supply Chain
- Toth, P., & Vigo, D. (2014). Vehicle Routing: Problems, Methods, and Applications (2nd ed.). SIAM.
- Golden, B.L., Raghavan, S., & Wasil, E.A. (2008). The Vehicle Routing Problem: Latest Advances and New Challenges. Springer.
- World Bank. (2023). Logistics Performance Index (LPI) 2023. World Bank Group.
- McKinsey & Company. (2024). The Future of Logistics: AI-Powered Transportation. McKinsey.
- Gartner. (2024). Top Transportation & Route Optimization Trends. Gartner Research.
- AurinoWorks. (2024). Transportation & Logistics in Indonesian Context: Case Studies & Best Practices. Internal Research.
📦 Materi Pelengkap / Arsip
Konten Existing Sesi 06
Materi berikut merupakan pelengkap pemahaman SCM secara menyeluruh. Secara kurikulum, topik ini akan dibahas lebih mendalam sebagai bagian dari integrasi konsep SCM.
Global Network Design: Mengelola Risiko & Ketidakpastian
Deskripsi Sesi: Jangan terjebak “harga murah” saat *offshoring*. Sesi ini mengajarkan cara menghitung Total Landed Cost dan menggunakan Decision Tree untuk mengambil keputusan investasi di bawah ketidakpastian.
Pendahuluan: The Offshoring Trap
Banyak manajer memindahkan pabrik ke negara bergaji rendah (Offshoring) hanya karena melihat “Unit Price” yang murah. Mereka lupa menghitung biaya tersembunyi: logistik, bea masuk, inventory in-transit, dan risiko kualitas.
“Persaingan bukan lagi antar perusahaan, tapi antar rantai pasok. Rantai pasok yang global memiliki peluang laba besar, tapi juga risiko yang eksponensial.”
1. Konsep Total Landed Cost (TLC)
Biaya barang hanyalah puncak gunung es. Untuk keputusan sourcing yang benar, hitunglah TLC:
1. Unit Price
Harga beli barang (FOB).
2. Logistics Cost
Freight (Laut/Udara), Asuransi, Handling pelabuhan.
3. Customs & Duties
Bea masuk, PPN Impor, Pajak barang mewah.
4. Inventory Cost (Hidden!)
Biaya modal mati selama barang di laut (4 minggu) + Safety stock ekstra.
5. Risk Cost (Hidden!)
Biaya keterlambatan, fluktuasi kurs, pencurian IP.
2. Manajemen Risiko Rantai Pasok
Risiko global bisa dikelompokkan menjadi tiga:
| Jenis Risiko | Contoh Kejadian | Strategi Mitigasi |
|---|---|---|
| Supply Risk | Pabrik vendor kebakaran, pemogokan buruh. | Dual-sourcing (Jangan taruh semua telur dalam satu keranjang). |
| Demand Risk | Perubahan selera pasar, ekonomi lesu. | Aggregation (Pusatkan stok di hub regional). |
| Exchange Rate Risk | Rupiah melemah terhadap Dolar. | Natural Hedging (Jual dan beli dalam mata uang yang sama). |
3. Decision Tree: Alat Pengambil Keputusan di Bawah Ketidakpastian
Bagaimana memutuskan investasi jangka panjang jika masa depan tidak pasti? Gunakan Decision Tree Analysis.
🧮 Studi Kasus: Onshore vs Offshore
Pilihan:
- Opsi A (Offshore – China): Harga unit murah, tapi harus kontrak volume besar (Kaku).
- Opsi B (Onshore – Lokal): Harga unit mahal, tapi boleh beli eceran (Fleksibel).
Ketidakpastian: Permintaan bisa TINGGI (50%) atau RENDAH (50%).
Analisis Skenario:
- Jika Demand RENDAH: Opsi A (China) akan rugi besar karena gudang penuh barang tak laku. Opsi B (Lokal) aman karena bisa stop beli.
- Jika Demand TINGGI: Opsi A (China) untung besar karena margin tebal. Opsi B (Lokal) untung tipis.
Kesimpulan NPV: Setelah dihitung dengan probabilitas, seringkali Opsi B (Fleksibel) memiliki nilai harapan (Expected Value) yang lebih tinggi karena risikonya terukur.
🎥 Tutorial: Menghitung Decision Tree di Excel
4. Strategi Hibrida (The Best of Both Worlds)
Perusahaan cerdas seperti Dell dan Apple tidak memilih salah satu, tapi keduanya:
Strategi Dell:
Gunakan komponen dasar murah dari Asia (Low Cost) via laut.
Tapi untuk produk baru/panas, gunakan kargo udara (High Speed).
Kombinasi Efisiensi + Responsivitas.
Ringkasan Eksekutif Sesi 06
- Hidden Costs: Jangan terbuai harga unit murah. Hitung Total Landed Cost.
- Value of Flexibility: Fleksibilitas (kemampuan mengubah volume) adalah aset berharga yang harus dihitung nilainya.
- Analytical Tool: Gunakan Decision Tree untuk memetakan skenario terbaik dan terburuk sebelum tanda tangan kontrak jangka panjang.
Referensi:
Chopra, S., & Meindl, P. (2016). Supply chain management: Strategy, planning, and operation (6th ed.). Pearson. (Bab 6: Designing Global Supply Chain Networks).
Chopra, S., & Meindl, P. (2016). Supply chain management: Strategy, planning, and operation (6th ed.). Pearson. (Bab 6: Designing Global Supply Chain Networks).