Date: 14th November 2018
Location: Data-Driven Ldn Theatre
Time: 11:10 - 11:40
Organisation: Microlise
About: Microlise are a leading provider of technology solutions to the transport and logistics industry worldwide. Discover how, with over 400,000 connected assets generating billions of messages a day, Microlise is evolving its platform to bring real-time analytics to its customers to improve safety, security and efficiency outcomes.
3. OUR MISSION:
TO CREATE COMPELLING INFORMATION
FROM CONNECTED MOBILE DEVICES
Microlise helps customers to reduce the costs and
environmental impact of their fleet operations.
We achieve this by maximising utilisation, increasing
efficiency, and improving economy and safety.
4. MICROLISE – THE HEADLINES
Pop-up: Pic of staff
TBC
UK HQ in Nottingham 450+ Employees £50m Revenues
400,000 Connections Worldwide Offices Internationally OEM Partners
5. FLEET PERFORMANCE
< BACK TO PRODUCTS
FLEET VISIBILITY &
ACTIVITY MONITORING >
DRIVER PERFORMANCE
>
DRIVETAB >
DRIVER & VEHICLE COMPLIANCE
>
SAFETY >
DRIVER COMMUNICATIONS
>
VEHICLE HEALTH >
Includes a range of standard and
optional modules, Fleet Performance
supports customers in maximising the
safety and efficiency of drivers and
vehicles in their fleets.
The product minimises environmental
impact and costs through improved fuel
economy, whilst maximising fleet
utilisation and efficiency.
CAMERA >
TRAILER TRACKING >
6. Journey Management offers
visibility of the plan vs its schedule,
facilitating proactive customer
service and issues management.
Improve planning through planned
vs actual comparisons.
< BACK TO PRODUCTS
JOURNEY MANAGEMENT
SCHEDULE MANAGEMENT >
CUSTOMER COMMUNICATIONS >
DRIVER PERFORMANCE >
ROUTE MANAGEMENT >
DRIVER COMMUNICATIONS >
WORKFORCE & RESOURCE MANAGEMENT
>
CUSTOMER RELATIONSHIP MANAGEMENT
(CRM) >
7. COLLECTION AND DELIVERY MANAGMENT
>
SUBCONTRACTOR VISIBILITY/SMARTPOD >
VEHICLE LOADING >
CUSTOMER RELATIONSHIP MANAGEMENT
(CRM) >
POD WORKFLOW BY SECTOR >
CUSTOMER COMMUNICATIONS >
< BACK TO PRODUCTS
EPOD
HARDWARE OPTIONS >
Increase accuracy, reduce
administration and go paperless
with Electronic Proof of Delivery.
8. WHY IS BIG DATA IMPORTANT TO US?
2017 stats - we capture lots of data!
12. OUR JOURNEY – THE BEGINNING
• Our journey began in 2016
• We collect terabytes of telematics data per month – we just didn’t have
the keys to unlock the full potential
• We also had many issues that we knew we were going to face in the
very near future
• There were also new technology drivers, as well as business drivers,
that presented us a great opportunity for change.
13. OUR JOURNEY – OUR BIGGEST ISSUES
• Data spread
– Thousands of silos, across hundreds of servers
• Data explosion
– More units
– More events
– More inputs
• Data variety
– Telematics
– Routes and Journeys
– Consignments
14. OUR JOURNEY – ASSESSING THE FUTURE
• Started small
– Built a few small test systems
– Assessed the tools
• We began to see that there was a considerable learning curve
• Started to build a picture of what a production ready solution was
– On Premise
• We realised that we need assistance
– We needed something that was supported
– Based our assessment on Use ability, supportability, security etc
15. OUR JOURNEY – BUILDING OUR FUTURE
• We used Cloudera Professional services as a kick start our production
environment
• We focus a lot of effort on training
– Cloudera onsite training
• Administration, Developer, Data Scientist
– Online courses such as Pluralsight.com
– Books
16. OUR JOURNEY – BUILDING OUR FUTURE
• We had a great project to build our production ready Proof of Concept, a government
funded project to collect telematics data from Low Emission Fleet Vehicles
– Real use case
– Time constrained
– Limited Scope
• This had the requirement of capturing telematics data, producing dashboards and reports
for a small number of vehicles, looking at new data elements that we had not captured
before.
• It would have taken us 6+ months in our existing solution to set this up to even start
capturing the data and longer to produce the reporting layers.
21. A secondary objective was to gain value from some of the data that we had stored on offline
storage
• Restored over 150 SQL Databases
• Used Sqoop to extract data and move into HDFS
• Spark to process the data into something we could use to get insights and value
OUR JOURNEY – WHAT WE ACHIEVED
23. OUR JOURNEY – CHALLENGES
Lack of knowledge
• Internal staff needed training
• Lack of skills on the market
• Use contractors for knowledge sharing
On Premise Installation
• Had to upskill in Linux admin
• Having to support all aspects of the solution
• Most online resource about the cloud
Business Level Scepticism
• Made it hard to get initial team
• Now its been proven - we are adopting the platform for more work
24. OUR JOURNEY – PENDING CHALLENGE
• Everyone needs it
• Nobody talks about it
• Not just for protecting against Data Centre failure
• Not cheap
• We have a plan, but HA is good enough for now
25. OUR JOURNEY – LESSONS LEARNED
We learnt many lessons along the way
• Schema evolution with JSON on a Kafka topic was difficult
• Data science teams use a lot of resources
• We used Spark where perhaps we didn’t need to
• More than one tool for the job – if struggling, try something
different
• Get professional advice
27. OUR BIG DATA STRATEGY
BIG DATA,
DATA SCIENCE &
ARTIFICIAL
INTELLIGENCE
Core Product Enhancement
Consultancy & Data
Vehicle Engineering Support
Future Proofing our Portfolio
Risk Management
Driver & Ops Benchmarking
Customer Insight Tools
AI Driven Decision Making
Machine learning and Optimisation
Risk Management
Traffic Flow
Marketing Support
Data Aggregation & Export
Vehicle Performance Analysis
Fault Analysis
Predictive Analytics
Multi-brand Diagnostics
Customer Insight Tools
Data aggregation & Export
V2X & ADAS Data Analytics
Autonomous Vehicle Learning
Platoon & Multi-modal Planning
28. ANALYTICS FROM DESCRIPTIVE TO PRESCRIPTIVE
Prescriptive
Predictive
Diagnostics
Descriptive
Optimisation What is the best that can happen
Predictive Modelling
Forecasting
Statistical Analysis
Alerts
Query Drilldown
What will happen next
What if these trends continue
Why is this happening
What actions are needed
Where exactly is the problem
How many, how often, where
What happened
Ad-Hoc Reports
Standard Reports
CompetitiveAdvantage/
DegreeofIntelligence/
AnalyticsMaturity&Capabilities
Levels of Analytics
10 years
ago
Now
Next year
5 years
ago
30. AI IN ROAD TRANSPORT
— Connected and Autonomous Vehicles
— Truck Platooning
— Road Sign Recognition
— Predicting Traffic demand
— Signal Control of Traffic at Road Intersections
— Dynamic Route Guidance
— Automatic Incident Detection
— Classes of Drivers based on Driver Behaviour
— Road and Driver Risk Prediction
— Predictive Maintenance
31. THE FUTURE FUTURE
Business Decisions Focus on Harsh Cornering
Supporting Data Anonymised Event Data
Historical Performance Data Consistent Harsh Cornering
Real Time Performance Data Journey Planning Information
Effects of Harsh Cornering
(2 minutes)
Effects of Harsh Cornering
(2 minutes)
How to reduce….
(3 minutes)
http://www.example.com
/harsh-cornering-best-
practices
Best Practices Blog
(4 minute read)