Planning for more resilient transport networks

Editor's Notes

• #6: We are interested in these late trips
• #7: Variability is a useful planning metric but it only represents reliability when considered over a specific range Reliability considers the variability of travel times compared to a historic ‘acceptable’ travel time The sample ‘looks back’ across a historic dataset In this sense it is multi-dimensional as it is consists of more than a single aggregate observation – It cannot be measured at a point in time unlike speed Reliability is a measure of unacceptable system performance
• #8: The metric cannot be observed at a point in time – requires a historical reference This is an example of a tardy trip metric such as the one employed by TfL and now MRWA and RMS. It is a system focussed metric Other metrics include buffer type (% additional time) – user focussed metric and statistical variation
• #10: SCENARIOS: 1) IDEAL 2) WORST CASE 3) & 4) EARLY DEPARTURES We all have values of time which are dependent on the type of activity we are undertaking… commuter and business activities have the associated highest values of time Transport improvements are targeted towards delivering improved journey times which allow us to get to destinations more quickly Appraisal of transport schemes is driven by econometric principles. A good transport systems allows us to get to destinations quickly meaning less lost, or unproductive time (spent travelling) and more time spending money or earning money both activities contribute to the economy… this is why transport is important Schemes that deliver the greatest benefits (primarily journey time) across all users, in consideration of value of time, are often the most viable (independent of political pressures etc, etc) There is now an increasing focus on JTR…. Cut to slide. Unreliable journeys contribute to an increase in unproductivity A PT trip has two components – Waiting and in vehicle travel-time
• #12: 2010 Transport User Analysis – Surveys SEQ transport users – Key considerations, 2/3 ranked reliability in top 3, ½ ranked shortest trip Studies in the USA state that between 1982 and 2001 average travel times increased by 20% whereas excess travel times increased by 25% Reliability is the most important aspect of service (Holland, 2003, bus-rail, tram) In service variability equates to roughly half of waiting time variability (MVA, 200, Bus Users in France) People typically value one minute late as 30% to 50% more than one minute of delay Women typically value being late up to 3 times as much as men (USA, 2002, State Route 92 Toll Route California) Non-recurrent delay constitutes up to 30% of the delay experienced on roads in our major cities Despite all this reliability is critically under-represented in transport scheme appraisal. Question of disentangling delay and reliability. A scheme that delivers travel time improvements will not always deliver reliability improvement. But the alternative could be argued to be true. Is JTR an apex statistic.
• #14: Non-recurrent congestion contributes to up to 30% of the delays we experience on our road transport networks and it is increasing beyond the rate of average traffic growth Bottlenecks /poor design and signals (recurrent factors) contribute to 45% of non-recurrent delay – design targeted towards maximising theoretical throughput under optimal system state departures seek to deliver reduces cost at elevated safety and reliability risk
• #15: Non-recurrant events typically generate between 10% and 15% of delay experienced in cities such as Brisbane and Sydney
• #16: Direct impacts Secondary impacts – cascading into adjacent and parallel transport networks – and not just limited to road Compounded impacts Secondary events
• #17: Individual behaviours and choices responses to changing traffic conditions Perfect knowledge and individual choice 90% of crashes are a result of human error
• #20: See non-recurrent events as risks to a reliable network Intensity is inherently linked to the current system state (how congested it is) – regime based scenario development Build an understanding of the different scenarios, influencers and barriers
• #21: Look at the policy chain – what do / can you affect – where is the source of your issues – is something systemic across non-recurrent events Build – Infrastructure, Technology assets, Business and performance systems, Governance structures – PINCHPOINTS Traditional approach = build -> manage -> price -> inform Operational and technology based approach = build/manage -> inform -> build -> price Shift in planning from building new roads (but we do need them) to managing what we have better Change in focus from delivering more supply to managing our demand
• #24: People typically value one minute late as 10% to 50% more than one minute of travel time Women typically value being late up to 3 times as much as men (USA, California)
• #25: Road authorities wish to deliver a road network which is resilient; stretches to manage major shocks and recovers quickly from them; and enables safe, efficient and consistent travel. To address these requirements, ITS implementation and the process of ITS design must follow the following principles:  Close coupling of ITS design with the civil design of roadways to ensure that the road may be actively managed in times of stress and is fit-for-purpose from the outset.  Consideration of the road network as a complete system, and understanding of impacts from changes to minor offshoots. This means that ITS tools must be deployed to operate at lane level across the system, enabling control of the entire network, major motorways, arterials and feeder roads.  Availability and use of real time data in control algorithms to accurately pin-point stress causal factors. Data density must be at a level which enables operational staff to determine mobility patterns over time.  Regular analysis of historical data to enable system optimisation and enhanced operational decision making. Additionally, the control algorithms should be able to perform predictive analysis in response to mobility patterns.  Deployment of high reliability and accurate devices and supporting infrastructure (including power and control systems).
• #26: Current trends in technology may be summarised thus: • Decreasing cost of connection due to increased data transmission speeds. • Change in acceptable base standard for electronic devices. More devices are being created with in-built sensors and Wi-Fi and 3G/4G connectivity. • Technology costs are decreasing, and smartphone penetration is sky-rocketing The net result is that data density is increasing. IOT allows us to take advantage of this increased data density facilitating the real-time collection and analysis of information from multiple sources, providing us with a holistic understanding of network performance. Infrastructure based technologies which dominate our transport networks (road, rail, freight, etc.) are traditionally closed systems, stored and managed in separate databases with a lack of inter-operability between technologies. IOT allows greater flexibility in device selection by promoting platform agnosticism. For example, data generated by a rail approved device can be used alongside information from a roadside unit, to enable a complete view of the transport network at any point in time. The IOT also allows operational staff to affect change more rapidly, and with greater coverage by pushing out information to personal devices. This allows users to make more informed choices and enables efficient implementation of the right operational strategy for current conditions (increased system agility).
• #28: Existing traffic monitoring systems are able to provide estimated travel time via detector data gathered from various points on the network. Research conducted by PB Consult (USA) and TRB suggests the use of a probability density function or cumulative distribution function (CDF) to characterize performance Additionally, it enables operational staff to be proactive in identifying issues on particular days and over specific routes. It provides objective data to enable a common roadmap for managing the network.
• #29: The future of driving is generally agreed to lie in the connected automated realm. When connected automated vehicles achieve sufficient penetration in the market, the likelihood of drivers acting as individual decision makers on the roads will be minimised, as cars interact with each other and the wider infrastructure, and take appropriate corrective action without any human input whatsoever. The integration of swarm intelligence technology in car design is already happening. The technology allows vehicles to collaborate and share data with others, thereby acting as a collective swarm for the benefit of the wider transport infrastructure (the cloud). Carmakers (Volvo, Mercedes and other members of the CAR 2 CAR Consortium in Europe) are actively developing systems for integration within future vehicles which use data garnered from the surrounding traffic environment to help reduce congestion and the number of accidents.
• #30: Change picture to temporary barriers in Delhi to support tidal traffic movements
• #32: Break an event down to its component parts Looks to address those component parts Consider a hierarchy of implementation and action / command structure Duration = Good communication (two-way) and capacity to respond (fleet and people) Intensity = Good decision making and available tools Extent = Management and control over broader network
• #33: Also reference devices and range of information
• #36: REDUCE TO KEY THEMES