Create an airport performance dashboard that reflects relevant KPI's. The dashboard should support Schiphol in becoming the smartest airport in the world, and reflect its throughput and its efficiency. Examples of KPI's are about traffic flows through Schiphol, or related to predictability of Schiphol (for Airlines and land-side). More extensive examples can be found in the detailed challenge description.
Potential users are customers of Schiphol, to see improvements over time as effect of measurements. The dashboard should be interactive and various view or data range selections should be possible.
Data that can be used is anonymised Cargonaut data (shared after signed NDA) to be enriched with public or social data.
Based on a number of DC locations and associated in/outbound numbers per time-slot per day, perform thorough data analysis and define improvement possibilities. Data to be used is data on inbound and outbound cargoes in combination with traffic state data (either historic or real time).
Problem definition: on certain highway bottlenecks (will be specified prior to June 1st) we experience recurring traffic jams. It is estimated that traffic jams generate a costs of 800 million euro annually, for transport only.
Most traffic forecasting models are based on data from the past. This challenge is based on real-time f traffic data, in combination with local weather data and road construction data.
Challenge question: is it possible te develop a forecasting model that will be able to forecast these upcoming traffic jams? How predictable are some traffic jams if we combine real time traffic data with other data sources like weather forecasts, event data, social data etc?
We all know that most trucks and white vans are only utilized for about 40-50%. What would the effect be is this can be raised to 60 or 70%? Is there a relation between the economic climate and modal split?
Try to visualize this by means of an app with dynamic controls, include the reduction in CO2 footprint and other factors.
Data on number of vehicles are available as public source as well as traffic density API's on main highways.
For the brave one's. Can we also turn this around; an economic barometer based on transport movements?
This is a research challenge. To what extend can we use social media or other social or public data to improve planning of consumer deliveries (as a result of exponential growth of online orders and deliveries) ? Which data sources seems the most suitable for this?
Nowadays most containers are planned for pick-up after ship departure. What if containers could be pick up at "estimated time of discharge"? Would this give the planner more time to go for multi-modal driven transports and thereby save money and reduce CO2 footprint?
Portbase will supply the related datasets for this challenge together with some need to know facts. You will need to combine this with other data like: intermodal connections and schedules, carbon footprint, etc.
The port of Rotterdam has many vessels entering and leaving the port each day. Besides the seagoing vessels an even bigger part of these vessels are related to inland (river going) barges. Interesting challenge is if there is any mutual relation between the sea and river going movements. Do the seagoing vessel (movements) affect the river going movements? Or the other way around. Is there any predictive behavior based on historical data etc. ?
For example: at the Maasvlakte area both sea and river going vessels are competing for the same terminal handling capacity. If a large number of seagoing vessels are in, less barges can be handled. As a result barges are confronted with serious delays (=additional costs).
Available data: the vessel (object) movements of the port and dedicated areas within the port (geofence) are tracked and traced and their data is logged. For the Maasvlakte area a data set with vessel object id's, time stamp, area code etc. is available. External sources which can also be used are vessel type data, weather forecasts etc.
Cancelled (no data available in time)
A major asset insurance company will provide data about their insured fleet. Data like: date, time, kind of damage, location etc. Challenge is to combine this data with traffic state data, weather data, age profile data etc. We are interested to know whether any correlations can be found and if transporters could be incentified to change routes or delivery windows in order to reduce damages?