PLATFORM ECOSYSTEMS FOR SMART CITIES: CASE STUDY IN INDONESIA

• Main Author: Mahesa, Restu
• Format: Dissertations
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/49386
• Institution: Institut Teknologi Bandung

Urbanization has brought challenges, coming from the limited capacity of the environment, and influencing social and economic aspects in urban areas. These urban challenges have motivated researchers to study the concept of urban sustainability. With the increase of information and communication technology, smart cities initiatives have grown in Europe, North America, and Asia starting in 2010. The Government of Indonesia has launched a program called "100 Indonesia Smart Cities Movement" in 2017, which targeted fifty smart cities in 2017, seventyfive smart cities in 2018 and one hundred smart cities in 2019, to create sustainable cities and support the improvement of citizen quality of life. This research aims to discover major constructs and proposing the model of platform ecosystems for Indonesia smart cities. This case study research applies the mixed methods research with an exploratory design approach that combines sequential qualitative and quantitative paradigms for model building and model testing. The literature review studies the previous researches of smart cities and collecting fourteen integrative smart cities frameworks that are gathered from global smart cities which are ranging in 2007 – 2017. The inductive qualitative phase in model building applied interview research techniques to identify current practices from the Indonesia smart city stakeholders. Ten interviews were conducted during data collection from September 2018 to February 2019. Five major constructs, namely Urban Systems, Political Leadership, Smart Urban Services, Citizen Preferences, and Policy Responsiveness, are connected with the literature review. One construct called Cultural Systems, which is unique and rarely mentioned in previous researches, has been revealed during this study and has been practiced by the Indonesia smart city stakeholders, is. The model building in the qualitative phase has suggested the hypothesized model, which comprises six major constructs, with 27 dimensions and 110 indicators with twelve hypotheses. ii Furthermore, the pilot test procedure was conducted using the Partial Least Squares Structural Equation Modelling (PLS-SEM) for questionnaire validation from July 2019 - August 2019. One hundred fifty-one respondents participated in the pilot test. The data analysis using Smart PLS software and the respecification process was performed. Some of the indicators and dimensions were eliminated and merged. The Average Variance Extracted (AVE) value ranging from 0.575 to 0.844, and Composite Reliability (CR) value ranging from 0.905 to 0.968, indicated good validity and reliability. The final result of the pilot test was a final questionnaire representing six major constructs, 26 dimensions, and 98 indicators, and ready for a full-scale questionnaire survey in large samples. The final survey was administered in six of Indonesia's major cities, namely Medan, Jakarta, Semarang, Surabaya, Makassar, and Samarinda, from October 2019 to November 2019. The cities are part of the "100 Indonesia Smart Cities Movement," and representing the major cities of four big Islands with high HDI (Human Development Index) in Indonesia. The data from 1021 respondents were gathered. It comprised of one hundred sixty-eight or more respondents in every city. Every subclass of stakeholders has forty or more respondents and fulfilling the minimum criteria of thirty respondents to hold the Central Limit Theorem of normal distribution samples. The study applies Model Generating (MG) strategy with the Covariance Based Structural Equation Modelling (CB-SEM) for model testing. The normality and outliers' tests suggest that the data are multivariate data distribution and shows no evidence of outliers. Confirmatory Factor Analysis (CFA) checks the validity of the measurement model. The CFA analysis shows a very good model fit, as shown by ?2/DF value of 2.339. Further examination shows no multicollinearity issues found. The respecification process using AMOS Software has been performed, and the good parameters of model fit were achieved, as shown in the Comparative Fit Index (CFI) value 0.903, The Tucker-Lewis Index (TLI) value of 0.900, The Root Mean Square Error of Approximation (RMSEA) value of 0.036. All indicators factor loadings value of 0.622 to 0.798. The results showed satisfactory reliability and validity. The structural model result analysis shows that all path analysis of constructs relationships satisfies high statistical significance. Further investigations of twelve hypotheses result in ten hypotheses are supported, and two hypotheses are rejected. The negative relationship estimate of Urban Systems to Smart Urban Services linkage has caused the rejection of H3 and H4 hypotheses. Based on more in-depth data investigation, we presume that there is a problem in Urban Systems aspect in cities that received negative perceptions from the respondent of several cities. The possible alternative explanations linked to underlying theories are offered. Triangulation procedure has been elaborated by comparing the qualitative and quantitative findings to the hypotheses investigation results. Policy implications and future research agenda are concluded. iii The research claims a new construct, Cultural Systems with its dimensions and indicators, which shows very good validity and reliability as a novelty to extend the body of urban sustainability knowledge. Keywords: platform ecosystems; smart cities; sustainable urban development; cultural systems