1. Analytics in Action
Case Study
07 February 2019
Transportation sector
http://DSign4.education/case_studies
Presented by,
Vivek Aradhya S U
Venukumar M
Group 4
3. Case Study Questions
Transportation sector:
Category of companies that provide services moving people, goods or
infrastructure.
AI usage in
various sector
4. Case Study Questions
Transportation sector:
Category of companies that provide services moving people, goods or
infrastructure.
AI usage in
Transport sector
5. DHL Business model
Key partners: Airlines, customers, e commerce, Manufacturers etc...
Key activities: Logistics
Value proposition: Low cost shipments, door to door shipments,
return shipments, tracking, imports exports.
Customer segments: Private and Business
Costs: Vehicle maintenance, fuel, Labour, R and D
Business Model
6. DHL Business model
Revenue stream: Monetizes by providing the delivery services like
• Providing air, surface or combined courier delivery services
• Providing express delivery services
• Delivering parcels, documents and packages
• Packing and sorting operations
• Transferring and trucking (without storage)
• Delivering time-sensitive and high-value-to-weight products
Business Model
7. Why organisation focus on data?
• We know exponential data growth but...
• What additional value does the existing bulk of data carry and how can we
capitalize on it?
• Change of attitude about how to use data – New experimentation, getting required
information, integrate all systems etc…
• Data-driven insight to achieve effective business decision making scenarios like…
Reason for focus on data
Operational efficiency:
• Increase level of transparency,
• Optimize resource consumptions, right staffing.
• Improve process quality and performance.
Customer experience
• Increase retention, loyalty, satisfaction from service.
• Perform precise STP (Segmenting targeting and positioning) by getting latent information.
• Personalise customer interaction service
8. Why organisation focus on data?
Reason for focus on data
What data?
Origin and destination, size, weight, content, location, trucks, smartphones,
RFID readers, Video footages, sensors, e-commerce logs, customer discussions,
voice of customers, previous transactions etc..
9. AI roadmap
Perception: Reliable and consistent
delivery with low cost
Prediction: NLP model, Machine learning, AI,
Forecasting, Warehouse management
Evaluation: Increase in revenue, customer loyalty and retention
Action: Taking better decision to improve operational efficiency and
process
10. Data science technique company used/favour?
Bottom up approach
Geographical mapping of trucks and technique of using historical data
to achieve pattern recognition and therefore predict ETA, shift
schedules in advance, resources to meet customer demand.
Results
Actions
Knowledge
Context
Data
Measures
Decisions
Interprets
ProcessObtain
Define
Require
Drive
11. Data science technique company used/favour?
Issue1:
Rely on highly complex manual processes that requires skilful knowledgeable staff.
Logistics is also an effort-intensive process. It is difficult for human workers to
maintain consistent levels of concentration throughout the workday. This can
result in costly mistakes.
12. Data science technique company used/favour?
AI in back office process:
• Learning from human decisions
• Making fast judgements
• Interacting with humans
System uses natural language processing to classify any clauses, policy sections,
and signature portions. Paired with a human to review these findings. Saves time.
Issue1:
13. Data science technique company used/favour?
Issue2: Shipment delays
By analysing different parameters of data, it is able to predict if the average daily
transit time for a given lane is expected to rise or fall up to a week in advance
DHL has developed a machine learning based
tool to predict air freight transit time delays.
Also identifies the top factors influencing shipment delays,
14. Data science technique company used/favour?
• Unexpected spikes in demand of Fidget spinner in 2017
predicted by AI, helped in predictive demand and capacity
Planning.
• AI predicted this demand from web data, online browsing data,
YouTube video views and conversations on social media.
15. Data science technique company used/favour?
Issue3: Continuous SCM
In various sectors managing the flow of components from thousands of worldwide
suppliers is a regular part of daily business.
The Resilience360 is a supply watch
monitor uses AI to mitigate supplier
risks.
Using advanced machine learning and natural language processing techniques, it
monitors the content and context of 8 million posts from over 300,000 online and
social media sources.
Problems with suppliers, from material shortages to poor labour can cause critical
disruptions in the supply chain.
16. Data science technique company used/favour?
Resilience360 understands sentiment of online conversations also identify risk
ahead of time.
This in turn allows supply chain managers to take corrective action earlier, and
avoid disruption.
17. Data science technique company used/favour?
Issue 4: The sorting of
unstructured damaged
parcels
The AI was trained on where to look in a given image and how to successfully
recognize and then classify into damage types.
Damage and wear to
shipments over time are
inevitable
18. Data science technique company used/favour?
From learning Watson’s visual recognition capabilities improved to
an accuracy rate of over 90% in just a short period of time.
20. What is the link between data science and decision makin
Data processing
using data scienceData
Information Data driven
decision
Leads to
• Quick response
• Minimum product damages
• Increase profits
• Great customer satisfaction
• Reliable and consistent delivery
• Increase trust
• Increase operational efficiency
• Increase satisfaction in 2 sided
market