A State-of-the-Art Review on Empirical Data Collection for External Governed Pedestrians Complex Movement

Table 2

SWOT analysis for laboratory experiments versus field observations.

Laboratory Experiments
Strengths	Weaknesses
(i) Specific dependent and independent variables can be controlled and examined one by one flexibly; (ii) Repeated measures can be realized easily; (iii) Save a lot of time comparing with field observation to capture the intended samples	(i) Discrepancy exists between real life situation; (ii) Uncontrolled latent variables may have effects on the global dynamics; (iii) Contextual limitations on the experiments targets and purposes
Opportunities	Threats
(i) The application of emerging data collection and processing methods can attract more researchers to perform experiments; (ii) Optimal experiments design can help lower the cost and save the time; (iii) A comprehensive empirical database can add on the values of current experimental data; (iv) Advanced statistical approaches might improve the performance of empirical data in verification and validation processes	(i) Organization and performance of experiments require a lot of money and labour; (ii) Data integration is hard due to the barriers among disciplines
Field Observations
Strengths	Weaknesses
(i) Ground truth data; (ii) Labour cost effective and easy to carry out; (iii) Large quantity of existing datasets	(i) Take long time to obtain adequate samples; (ii) Not easy to capture specific behavior compared with laboratory experiments; (iii) Restrictions of access into certain areas or contexts
Opportunities	Threats
(i) Smart city can provide vast data source for collecting pedestrian movement behaviors; (ii) Big data solutions might be able to reduce the need of data quality; (iii) Advancement in image processing industry offer the chance to collect large scale field data	(i) Complex patterns and behaviors are difficult to recognize and capture from field data; (ii) Privacy & security issues should be highlighted as the potential large scale in data volume; (iii) Minimize the effects of the placement of cameras or UAVs on the investigating subjects