Book Read Free

Digital Marketplaces Unleashed

Page 63

by Claudia Linnhoff-Popien


  In order to detect deviations from the predicted trend, a control chart was developed which illustrates the deviation of the predicted retail revenue from the actual revenue over time. It includes lower and upper boundaries for the deviations which trigger a warning whenever they are crossed. These warnings subsequently lead to a root cause analysis of the deviation and enable the airport management to take corresponding actions.

  40.3.2 Optimization of Aircraft Positioning

  The Smart Data Lab recently investigated the impact of aircraft stand allocation on retail revenue. Previous analysis had exposed that different passenger groups, e. g. passengers to certain destinations, have different shopping preferences and behaviors. The resultant challenge for the Smart Data Lab was then to analyze how the planned allocation of aircraft to stands on the aprons – which in turn, at least to a large extent, determines the path of the passengers through the terminals – can influence retail revenue.

  As a first step, a rather simple model only relying on gate information was considered, but this model had limited explanatory power only. Consequently, as a second step, a multivariate model taking into account gate, airline, destination, waiting time, dwelling time and other parameters was implemented and led to significant results. Based on these insights, a prototype was developed which allows to simulate effect on retail revenues of holistic positioning plans. Furthermore, the prototype is able to calculate the opportunity cost of suboptimal positioning plans and to suggest ‘retail‐optimized’ positions for connecting flights.

  40.3.3 Improvement of Estimated In‐Block Time

  Efficient planning of the ground handling activities at an airport requires precise estimations of the arrival times for incoming aircraft. An aircraft arriving at the parking position before the ground handing crew arrives on site may lead to fines to be paid by the ground handling department to the aircraft operator. On the other hand, if the aircraft arrives on stand a long time after the ground handling crew, they will be idle for some time, leading to inefficient resource allocation.

  Currently, the time stamp “Ten Minutes Out” (TMO), i. e. the aircraft is expected to land (touch down) in ten minutes, is used as a major triggering milestone for ground handling operations. By definition, TMO does not take into account the taxi‐in process of the aircraft from vacating the landing runway to the parking position. In the past, the taxi‐in time was taken from a fixed tabulation providing standard taxi times based on the planned runway and the parking position only.

  A Smart Data Lab project team analyzed the current situation and discovered that the majority of asynchronous arrivals of aircraft and ground handling crews at parking positions was due to unexpected high variations of the actual taxi‐in times from the standard taxi time. The main causes of this effect are comprehensible with common sense: the traffic at Frankfurt Airport has been growing significantly over the last decade and the complexity in surface traffic management has been increasing simultaneously and especially since the new runway 25 R/07 L opened in 2011 leading to converging traffic flows on the airport surface. However, the majority of changes in the taxi route leading to unexpected taxi‐in times is caused by runway changes during final approach, i. e. after TMO. The actual landing runway has a significant effect on the expected taxi time as the distances from the different landing runways to the allocated parking position may be very different.

  An alternative and promising time stamp in the aircraft arrival process is the “Estimated In‐Block Time” (EIBT) – the estimated time an arriving aircraft reaches its aircraft type specific stopping point on the parking position – including the taxi‐in process. A high quality EIBT available 15 or 20 min prior to the real in‐blocks event would be very helpful, since it would provide the ground handling crews with unprecedented planning reliability. Hence, the Smart Data Lab accepted the challenge to improve the EIBT calculation in a time range of approx. 15 min in advance of the actual event. As a first step, the team introduced a new time stamp called ‘15 min before On Block’(EIBT‐15). In order to establish and to calibrate an adequate calculation model delivering the new time stamp in sufficiently high quality, the interdisciplinary working group used historical aircraft surveillance data of the final approach and the taxi‐in phase.

  Finally, the Smart Data Lab project team was also able to quantify the gain in accuracy of the new EIBT‐15 time stamp compared to the currently used TMO plus fixed taxi‐in time. Based on this information, the business case of a system wide change from TMO to EIBT‐15 can be calculated: while the estimated gain in efficiency and reduction of penalties are the profits, CAPEX4 and OPEX5 for the change of various IT systems and procedures constitute the costs.

  This example reflects the rationale behind the Smart Data Lab: the lab itself can demonstrate a possible improvement based on statistical information. Nevertheless, the decision to use this improvement in daily business can only be made based on a positive business case for Fraport AG as a whole, taking into account all cost and benefits.

  40.4 Summary

  The example of the Passenger Flow Analysis shows the potentials of predictive analytics for the operational, and indirectly also the commercial, further improvement of Frankfurt Airport. Based on this insight, Fraport AG has developed the organizational and methodical approach of the Smart Data Lab. It aims at exploring comparable potentials systematically all across the Fraport business units. The examples dealing with airport retail revenue trends and improvements by optimal aircraft positioning, and also the improved ground handling efficiency due to the implementation of EIBT‐15, show how flexibly predictive analytics can be applied to various business scenarios – always assuming that a sound basis of historical data is available and powerful tools exist to explore them.

  Fraport AG will reiterate the Smart Data Lab initiative annually and with the clear objective to improve the business step by step making Frankfurt Airport a Hub Airport 4.0.

  References

  1.

  R. Felkel and D. Klann, “Comprehensive passenger flow management at Frankfirt Airport,” Journal of Airport Management, vol. 6, no. 2, pp. 107–124, 2012.

  2.

  ACI World Airport IT Standing Committee, “Best Practice on Automated Passenger Flow Measurement Solutions,” 27 August 2015. [Online]. Available: http://​www.​aci.​aero/​media/​2c9abef1-ae58-40ed-9d33-4bdd925aee89/​NGQA3Q/​About%20​ACI/​Priorities/​Facilitation/​Best-Practice-on-Automated-Passenger-Flow-Measurement-Solutions.​pdf.

  3.

  C. Mayer, R. Felkel and K. Peterson, “Best practice on automated passenger flow measurement solutions,” Journal of Airport Management, vol. 9, no. 2, pp. 144–153, Winter 2014–15.

  4.

  “Google Play – FRA App,” Goolge, [Online]. Available: https://​play.​google.​com/​store/​apps/​details?​id=​com.​infsoft.​android.​fraapp&​hl=​de. [Accessed 09 08 2016].

  5.

  “iTunes AppStore – FRA App,” Apple, [Online]. Available: https://​itunes.​apple.​com/​de/​app/​frankfurt-airport-fra-airport/​id453191399?​mt=​8. [Accessed 09 08 2016].

  Footnotes

  1Airports Council International World.

  2The arrival profile shows how many passengers typically show up at a process point in a predefined time pattern.

  3Human Machine Interface.

  4Capital Expenditure.

  5Operational Expenditure.

  Part X

  Mobility Services

  © Springer-Verlag GmbH Germany 2018

  Claudia Linnhoff-Popien, Ralf Schneider and Michael Zaddach (eds.)Digital Marketplaces Unleashedhttps://doi.org/10.1007/978-3-662-49275-8_41

  41. Preface: Beyond the Hood: the Development of Mobility Services in the Mobile Internet

  Markus Heyn1

&n
bsp; (1)BOSCH, Gerlingen, Germany

  41.1 Parking Spaces are Going Online

  Typically, any trip in a car ends up at a parking space. Of course, the driver has to find one first. In downtown areas, the search for parking spaces is responsible for roughly one‐third of traffic. Pressure on parking is growing, and curbside spaces are especially rare. Searching for an empty parking space is inconvenient, and usually time‐consuming and stressful. No wonder that looking for a parking space ranks in tenth place of the greatest worries of German car drivers1 and that, according to the online portal Statista, 87 % of drivers are interested in solutions that make it easier to find parking. That is why web‐based parking services have gained considerably in importance in recent years. With what is referred to as “cellphone parking,” drivers get a parking ticket from a machine when they drive into the parking lot but without paying. When they leave the parking lot, they send a text message with the parking number shown on their parking ticket to a number also shown on the ticket. The parking charges are then either deducted from their prepaid card or debited from their cellphone bill.

  But even when cellphone parking, driver still have to find a parking space. On average people take ten minutes and cover around 4.5 km when looking for somewhere to park, resulting in vehicle costs of 1.35 € per search and CO2 emissions of 1.3 kg per km2. Connectivity now makes it possible for vehicles themselves to find a free curbside parking space. With community‐based parking, the car becomes a driving parking sensor in the Internet of Things (IoT). It detects parking spaces on the side of the road with the sensors of its parking assistant. In Germany, the most common assistance systems in modern cars are parking assistants. According to a Bosch evaluation of the 2014 vehicle registration statistics, of the nearly three million cars that were registered that year, half of them (52 %) feature just such a system. The picture is fairly similar in other countries: in Belgium and the Netherlands, half of all new cars in 2014 (50 %) come equipped with a parking assistant. In the U.K., the figure is 19 %. Vehicles fitted with a parking assistant can recognize curbside spaces between parked cars as they drive past. The information is sent to the respective vehicle manufacturer by means of a communication interface and then forwarded in anonymized form to a cloud. Here, an intelligent process pools the data from all participating vehicles independently from thein brands to generate a digital parking map that is delivered back to the vehicle manufacturers. They in turn can share the map with all of their cars that are connected to the internet, for example via the navigation system. Drivers can then navigate straight to an available parking space. The search for a parking space could conceivably also be made less stressful by means of a mobile app. But the fact that not every curbside gap a car detects and reports automatically qualifies as a valid parking spot makes preparing the data a challenge. The gap could just as easily be a driveway, a bus stop, or a no‐parking zone. Data mining methods can be used to identify gaps next to the curb unequivocally as parking spaces. Should several vehicles repeatedly report a curbside gap as unoccupied, it is most likely not a valid parking space. Accordingly, these gaps are then not labeled as parking spaces on the digital parking map. The more vehicles that participate in community‐based parking, the more accurate and more comprehensive the service is. Once a certain number of users are participating, the digital finder for parking spaces can even provide information on a space’s length and width. This makes it possible to search for spaces that fit a specific vehicle, for instance a motorhome or compact car, in addition to the general benefit of considerably shortening the overall search for a parking space and lessening its environmental impact in cities. To be able to offer the service throughout Germany, it is necessary to work with multiple automakers. That is why community‐based parking was set up to be an open service platform in which multiple vehicle manufacturers can participate at the same time. The platform brings together a large number of participants and combines an offline parking offering with an innovative online service.

  Another way of ending what is at times a nerve‐wracking search for an empty parking space, especially in city centers, is an active parking lot management. This involves an intelligent technology detecting and reporting how many and which parking spaces are empty in a city center. These solutions are based on micromechanical components that are web‐enabled and installed in special occupancy sensors, which are subsequently discreetly placed in parking areas. The occupancy sensors are barely larger than a CD in terms of circumference and measure around three centimeters in height. They are installed in parking garages and on‐street parking spaces, either on or in the asphalt, as desired. Since they are battery‐powered and communicate using radio, there is no complicated laying of cables. The sensors check at regular intervals whether a parking space is occupied or not. Using an internet connection, they relay the securely encrypted information to a cloud. There, a real‐time parking map of all free and occupied spaces is created that can be accessed with the app or on the internet. Meta‐information on each parking space is available as well, including whether it is a space reserved for families, women, or the disabled, what the parking time costs, and if a charge spot for electric vehicles is available. Drivers can have themselves guided straight to the parking space by a smartphone app. More comprehensive services are also possible, such as a payment function by app. Yet active parking lot management has further advantages for parking lot operators, particularly as it can further improve the occupancy rate of heavily used parking spaces. The key to this is intelligent data evaluation. A web portal provides parking lot operators with a clear overview of which parking spaces were occupied by how many vehicles and when. During peak times, this information can direct drivers to spaces that are less frequently occupied, for example. In the Stuttgart region, Bosch is conducting further development work on active parking lot management in a pilot project with the Verband Region Stuttgart. The main idea behind it is that if drivers know they can find a free park‐and‐ride space, they will be more willing to use the city trains. Sensors fitted in 15 park‐and‐ride facilities along two city train lines will detect whether parking spaces are available or occupied. They will report this up‐to‐the‐minute information in real time to the Stuttgart transportation authority, VVS, which will make it available through its app and website. Eleven towns and municipalities in the northeast of the Stuttgart region have declared their willingness to support the pilot project. They will provide internet access and power connections for what are mostly municipal park‐and‐ride facilities. The Verband Region Stuttgart is supporting this project with a grant from the “Sustainable Model Region Stuttgart” state program.

  41.2 Intermodality Made Easy: Connected Mobility is the Better Mobility

  Many people, particularly in big cities, use more than one modes of transportation to get from A to B. They use buses, trams, trains, and car‐sharing – depending on whatever is suitable for their current need for mobility, what times these run, and what the current traffic situation is like. At times it has to be fast, at times comfortable, and at times particularly affordable. This is where integrated and digital mobility platforms come into play. Users can first obtain information on door‐to‐door mobility chains, then reserve or book them and also pay for them. Examples include a Daimler subsidiary’s moovel and Deutsche Bahn’s Qixxit. These apps see themselves as intermodal travel planners and intermediaries between providers and users of mobility services. Users enter the starting point and destination into the apps, then select the combination they personally prefer from a list of suitable means of transport. Separate contracts are concluded between the user and the providers of the relevant services. “Smile – einfach mobil” is an integrated mobility platform that promises a broader, more strongly integrated and more flexible range. It was developed i
n a joint project between the Austrian Federal Railways, Wiener Stadtwerken, and further participants. Smile intends to provide users with a comprehensive mobility range that they can use to conveniently combine mobility with public transportation services. A special focus is placed on integrating electro mobility services, such as electro car‐sharing, and using public charging infrastructure for electric vehicles used by their owners3. A new mobility assistant in the greater Stuttgart area is also acting on the idea of the innovative, intermodal connectivity of individual methods of transportation. It enables drivers to navigate routes using different modes of transportation, including bikes, trams and trains, buses, and sharing offers. It takes just one app to plan, book, and pay for a journey involving different modes of transportation. The mobility assistant for Stuttgart is supplied with real‐time data from a big data platform. It accompanies road users live via an app and takes account of any disturbances or obstacles, for example traffic jams or train delays. In such cases, it suggests better ways of reaching the destination more quickly and conveniently. Drivers can also use the assistant to book parking spaces. No matter the modes of transportation, at the end of the month the user receives one easy‐to‐read bill that covers all the mobility services used. In addition, the system has an interface and provides support to the Stuttgart region’s traffic management authority. For example, the assistant can be fed route recommendations that help manage traffic effectively and thus improve the traffic situation in the Stuttgart metropolitan area. Led by Bosch, the mobility assistant project is being implemented together with other urban mobility companies. Bosch is creating the central service platform, the intermodal navigation system, and the smartphone app used to operate the mobility assistant.

 

‹ Prev