透過您的圖書館登入
IP:3.22.100.180
  • 學位論文

探討某縣市運用商業智慧系統建構視覺化資料分析菸害防制稽查熱點

Explorating the utilization of business intelligence and visual data analysis on hot spot inspection for tobacco hazards prevention in a metropolitan

指導教授 : 郭鐘隆
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


背景:某院轄市為落實菸害防制執法成效,每年依據菸害防制法規範之室內及戶外禁止吸菸場所訂定稽查目標數,派稽查人員依規劃期程進行稽查,惟對於民眾陳情、民意單位反映、媒體報導等菸害案件,無法於第一時間蒐集違規吸菸熱點。本研究目的運用商業智慧系統探勘與菸害相關之衛生單位部門網路系統資訊,將大數據資料整合以視覺化資料分析禁止吸菸場所熱點分布,縮短室內及戶外禁菸熱點產出工作天數,並蒐集與違規吸菸熱點之關聯性,適時修訂菸害防制稽查工作,擬定防制策略,進而強化稽查效率、降低檢舉案件及提升行政效率。 方法:本研究以安裝Power BI 增益集的方式為開發平臺,使用Power Query將資料從陳情系統、派案系統、裁罰系統、稽查系統之數據經由萃取、轉置、篩選、正規化、載入方式至Excel資料表,再以Power Pivot載入為資料模型 (data model)、設定資料模型之間的關聯性,以進行數據的分析與彙算,最終以Power View互動式視覺效果 (visualization) 最終呈現。 結果:藉由商業智慧系統(Power BI),並使用Power Query將資料從陳情系統、派案系統、裁罰系統、稽查系統之資料庫數據分析發現,所佔比例最多之違規吸菸熱點行政區,分別為G區、A區最多,其次為D區、E區,第3為J區;違規吸菸熱點之場域,以公園為最多(主要為艋O公園),其次為休閒娛樂及提供民眾消費場所,第3為戶外公告禁止吸菸場所(香O廣場、捷運西O站外公告禁菸場所)之室內禁止吸菸場所為最多。另透過Power Query導入後,使操作步驟精簡化,稽查人員僅須更新資料來源與下達重新整理指令,使資料處理模式進入自動化運算流程,立即進行資料清冊產出,以直接檢視室內及戶外禁止吸菸場所熱點清冊,並派稽查人員至禁止吸菸場所進行菸害稽查,故可預期導入前後菸害熱點產出工作作業時間由4天縮短天數為1天,節省產出工作天數達75%;處理步驟由15個步驟簡化至5個步驟,簡化操作流程達67%,有顯著成效。 結論:運用商業智慧(Business Intelligence,BI)架構分析某院轄市菸害防制稽查熱點,重要研究發現,無論是系統面、稽查人員、決策人員等面向皆能提升稽查效能,減少重複檢舉發生,進而降低民眾陳情,未來具有引進全國,並提供各縣市執行菸害稽查業務之參考。

並列摘要


Background: In order to implement the effect of law enforcement on tobacco harm control, a city under the jurisdiction of a hospital sets the number of inspection targets for indoor and outdoor smoking bans regulated by the Tobacco Hazard Control Law every year. It is impossible to collect illegal smoking hotspots in the first time for cases of smoking harm such as petitions, public opinion units, and media reports. The purpose of this research is to use the business intelligence system to explore the network system information of health units and departments related to tobacco damage, integrate big data data to analyze the distribution of hot spots in non-smoking places with visual data, and shorten the number of working days for indoor and outdoor non-smoking hot spots. And collect the correlation with illegal smoking hotspots, revise the smoke prevention inspection work in a timely manner, formulate prevention strategies, and then strengthen the inspection efficiency, reduce the reported cases and improve the administrative efficiency. Method: This study uses the installation of Power BI gain set as the development platform, and uses Power Query to extract, transpose, filter, normalize and load the data from the reporting system, dispatch system, fee system, and audit system. Then use Power Pivot to load it as a data model, set the correlation between the data models for data analysis and calculation, and finally use Power View interactive visual effects (visualization) render. Result: Using the business intelligence system (Power BI) and using Power Query to analyze the data from the database data of the reporting system, the dispatch system, the penalty system, and the inspection system, it was found that the administrative areas with the largest proportion of illegal smoking hot spots, Areas G and A are the most, followed by Areas D and E, and the third is Area J; the most illegal smoking hotspots are parks (mainly Meng O Park), followed by leisure and entertainment and providing public services Among the consumption places, the third is the outdoor no-smoking places (Xiang O Plaza, the no-smoking places announced outside the MRT West O Station) with the largest number of indoor no-smoking places. In addition, after importing through Power Query, the operation steps are simplified, and the inspector only needs to update the data source and issue a rearrangement command, so that the data processing mode enters the automatic calculation process, and the data inventory output is immediately performed to directly inspect the indoor and outdoor prohibited smoking. Inventory of hot spots in smoking places, and dispatching inspectors to places where smoking is prohibited to conduct smoke hazard inspections, so it can be expected that the working time for the production of tobacco hazard hot spots before and after the introduction will be shortened from 4 days to 1 day, and the number of output working days will be saved by 75%; The processing steps have been simplified from 15 steps to 5 steps, and the operation process has been simplified by 67%, with remarkable results.. Conclusion: Using the business intelligence (BI) framework to analyze the hot spots of tobacco control and inspection in a city under the jurisdiction of a hospital, important research found that whether it is the system, inspectors, decision-makers and other aspects can improve the inspection efficiency and reduce the occurrence of repeated reports In order to reduce the public's complaints, it will be introduced to the whole country in the future, and provide reference for the implementation of smoke inspection business in counties and cities.

參考文獻


一、中文部分
王茁(2005)。商業智慧(初版)。新北市:博碩文化股份有限公司。
李玉秀、莫介中(2006)。逐步提昇企業智商,微軟「應用平台架構優化」研討會。台北福華。
李朝祥、林天生、溫翊伶(2015)。應用商業智慧建構海運公司決策戰情中心。蘭陽學報,(14),68-77。
李瑋倫、謝秉融、洪作緒、李沅銘、丁信修、陳大麟(2019)。人工智慧與大數據在油氣探採之應用。石油季刊,55(4),55-69。

延伸閱讀