Improved methods to deduct trip legs and mode from travel surveys using wearable GPS devices: A case study from the Greater Copenhagen area

Transportation mode detection – an in-depth review of applicability and reliability

“…Most research on mode detection in TSc relies on external GIS datasets. The reported accuracies vary: 91.7% accuracy for 4 travel modes (Chung & Shalaby, 2005), 94% accuracy for 7 travel modes (Tsui & Shalaby, 2006), 95% accuracy for 5 modes (Stopher et al, 2008), 83% accuracy for 5 modes (Schüssler et al, 2011), 85.8% accuracy for 7 modes, 70% accuracy for 6 modes (Bohte & Maat, 2009), 91.6% accuracy for 10 modes (Biljecki et al, 2013), 84.6% accuracy for 5 modes (Rasmussen et al, 2015), and 75.3% accuracy and 73.2% recall for 9 modes (Prelipcean et al, 2016).…”

“…Since there are no promptness constraints, TSc and HG approaches rely on external data sources such as road network datasets (Chung & Shalaby, 2005;Tsui & Shalaby, 2006;Stopher et al, 2008;Biljecki et al, 2013;Rasmussen et al, 2013Rasmussen et al, , 2015 or POI buildings datasets (Alvares et al, 2007;Palma et al, 2008). A direct consequence of the difference in focus between disciplines is the types of devices that are used to collect data and / or perform the mode detection.…”

“…In the body of research on mode detection in TSc, there seems to be a preference for inferring travel mode by using mostly subjective methods such as Fuzzy Logic (Tsui & Shalaby, 2006;Schüssler & Axhausen, 2009;Schüssler et al, 2011;Rasmussen et al, 2015) or any variation of fuzzy logic approaches such as methods based on Membership Functions (Biljecki et al, 2013), and Rule Based Heuristics (Chung & Shalaby, 2005;Bohte & Maat, 2009). Some methods also rely on objective methods such as Decision Trees (Zheng et al, 2010) or Random Forests (Montini et al, 2014), but these approaches constitute the minority.…”

“…A direct consequence of the difference in focus between disciplines is the types of devices that are used to collect data and / or perform the mode detection. While there is a prevalence for using smartphones in LBS (Manzoni et al, 2010;Reddy et al, 2010;Wang et al, 2010;Hemminki et al, 2013;Montini et al, 2014;Shah et al, 2014;Yu et al, 2014), TSc and HG mostly use dedicated devices such as the Geostats Geologger (Chung & Shalaby, 2005;Tsui & Shalaby, 2006) or other types of dedicated devices (Alvares et al, 2007;Palma et al, 2008;Stopher et al, 2008;Bohte & Maat, 2009;Schönfelder et al, 2002;Rocha et al, 2010;Zheng et al, 2010;Biljecki et al, 2013;Rasmussen et al, 2013Rasmussen et al, , 2015 with a few exceptions that use smartphones (Schüssler et al, 2011;Montini et al, 2014). This is an important distinction because, with a few exceptions (Stenneth et al, 2011;Shah et al, 2014), the studies that collect data using smartphones have a thick-client architecture, in which the client performs most inference operations and the server acts only as a storage backup, which contrasts those that use dedicated devices, which use a thin-client architecture, in which the server performs inference operations in addition to the data management.…”

“…Another widely used approach for segmentation is the one proposed by Zheng et al, 2010, which relies on the same assumption as Tsui & Shalaby, 2006, namely that a traveler walks when changing transportation modes, and detects walking segments by using a loose upper bound for velocity and acceleration (2.5 m/s for velocity, and 1.5 m/s 2 for acceleration). When analyzing the travel modes, there is usually a baseline for detected modes such as walking, bicycling and motorized (Chung & Shalaby, 2005;Tsui & Shalaby, 2006;Stopher et al, 2008;Schüssler & Axhausen, 2009;Schüssler et al, 2011;Montini, Rieser-Schüssler, & Axhausen, 2014;Bohte & Maat, 2009;Biljecki et al, 2013;Zheng et al, 2010;Rasmussen, Ingvardson, Halldórsdóttir, & Nielsen, 2013Prelipcean, Gidófalvi, & Susilo, 2016), but some research goes a step further and classifies the motorized into car, train, and public transportation (Stopher et al, 2008;Schüssler & Axhausen, 2009;Schüssler et al, 2011;Bohte & Maat, 2009;Biljecki et al, 2013;Rasmussen et al, 2015;Prelipcean et al, , 2016. Of these papers, Biljecki et al, 2013 provides the most detailed classification scheme that includes 10 modes: walking, bicycling, car, ferry boat, sail boat, train, subway, bus, tram and flight.…”