聯合國在2019年將塑膠汙染列為僅次於氣候變遷的危害，在環境中掩埋，洩漏在海洋的塑膠廢棄物已是全球災難。本研究結果不在於提倡環保，或綠色消費，畢竟大部份使用者在對於生活享受或產品樂趣是高於其為環境改變而減少享受等這類道德動機。本研究會聚焦在工業設計第一階段田野觀察，透過導入AI深度學習技術（Deep Learning），提高實驗大規模田野、取得更接近母體數量的樣本的可能性，來進一步精確了解，設計產品對於整個社會與使用習慣的影響，而驅動本文的研究動機。本研究希望提出更透明化、一致性、準確性的生活數據，透過科技力的輔助，來觀察個人生活空間（personal living space, PLS）裡的塑膠足跡的流動與使用習慣。鼓勵發展生態上有效的技術，以及大量的重新設計都市的循環基礎建設，並從中找到新未來的商業模型與產品開發決策因素。
本研究主要參考於ISO/TS14067 及PAS2050的盤查數據收集五大原則，以YOLO深度學習神經網路的辨識技術來建構塑膠足跡（Plastic Footprint），透過使用者提供生活空間照片進行間接式田野觀察，探討人類、環境、物件三者間的互動關係，提供工業設計師、決策者在跨部門設計溝通，評估永續產品開發對環境與社會影響的一套流程或介面。
研究方法 : 間接式田野觀察、機器學習、企業深度訪談。在研究成果部分，（1）AI辨識速度與準確率的驗證。本研究主要以機人合作概念，以深度學習辨識為主，人眼為輔，採YOLOv4深度學習神經網路進行塑膠物件的辨識，Mask R-CNN我們應用在前測階段負責標註塑膠產品，輔助人眼識別。在YOLO物件辨識的部分，經人工交叉比對驗證，其準確度可達92.55%；YOLOv4與YOLOv3相比，可增加23.76%（183個）的物件偵測命中率；照片資料分析與框選速度可較人眼（194.9sec / photo）提高兩百九十五倍到千倍之多（0.18~0.66sec / photo）；本研究樣本78人198張照片，跨越三縣市台南、高雄、雙北的26個行政區，四種不同的個人生活空間（PLS），房間、客廳、浴室、辦公室。而AI神經網路在照片中識別出的953個塑膠物件只用了130秒。並且經過人眼反覆確認，驗證只需要一名專業人員，約為數小時，保守估計實際應用可提高田野觀察的分析速度50倍以上，且達高準確度，適合大規模量化調查研究使用。（2）田野調查流程的簡化透由AI技術導入，傳統九大流程可縮減為四大流程。（3）建構數據視覺化，塑膠足跡與產品影響力地圖。（4）在驗證的部分，數據分析在箱型圖（Box Plot）顯示一個人在生活中持有物品的狀態，backpack、clock、keyboard、laptop、mouse、suitcase（行李箱），一個人擁有一件；chair、sofa、TV & monitor擁有1-2個；cup、teddy bear（玩偶），擁有1-3個；遙控器擁有2-3 個；potted plant（盆栽） 擁有2-4個；瓶罐類4-7個，有著極大離峰值（outlier）40、21、16件，每個人在此類有極高的變異（Variance）。實驗結果非常接近我們現實生活的體驗。（5）最後，本研究提出4種工業設計師或企業應用決策評估情境模擬，以供未來進行可建立在資料驅動上（Data-Driven）的設計。

This study primarily uses AI deep learning technology to identify photos to construct the footprints of plastic products in personal living spaces (PLS). We seek to determine whether its accuracy, reliability, and efficiency are close to that of human judgment and whether it can simplify the process and increase willingness to engage in fieldwork. Finally, the verified plastic footprint data is used to construct a plastic footprint product influence/impact map and related charts through data visualization, thereby constructing a process or tool to assist in sustainable product development decision-making. This research is primarily based on the five principles of ISO/TS14067 and PAS2050 inventory data collection. We use YOLO artificial intelligence identification technology to construct plastic footprints through a sample of 78 people with 198 photos, encompassing 26 districts in three counties, Tainan, Kaohsiung, Taipei, and four PLS—bedrooms, living rooms, bathrooms, and offices. Identifying the 953 plastic objects by the AI neural network in photos only took 130 seconds. These objects were repeatedly confirmed by the human eye, needing only one professional for the verification, which took a few hours. The interaction between humans, the environment, and objects was explored, industrial designers, managers, and executives were connected in cross-departmental research, and the development of sustainable products and the impact of their sets of processes and interfaces on the environment was evaluated.

外文文獻
Anton, V. (2019). Detecting plastic litter in natural environments. Retrieved from https://www.wildlife.ai/detecting-plastic-litter-in-natural-environments/
Biermann, L., Clewley, D., Martinez-Vicente, V., & Topouzelis, K. (2020). Finding Plastic Patches in Coastal Waters using Optical Satellite Data. Natural-Scientific Reports, 10(1), 5364.
BreakFreeFromPlastic (2018). Branded in Search of the World’s Top Corporate Plastic Polluters.
Brundtland, G. H. (1987). Our common future—Call for action. Environmental Conservation, 14(4), 291-294.
Burall, P.(1994). Green-ness is good for you. Design-London(544), 22-25.
C2CPII (2021). Cradle to Cradle Certified. Retrieved from https://www.c2ccertified.org/get-certified/product-certification
Castillo, J. J. (2009). Snowball sampling. Retrieved from https://explorable.com/snowball-sampling
Charter, M. (2018). Designing for the Circular Economy. UK: Routledge.
CIEL (2019). Plastic & Climate: The Hidden Costs of a Plastic Planet. US: the Center for International Environmental Law. Retrieved from https://www.ciel.org/wp-content/uploads/2019/05/Plastic-and-Climate-FINAL-2019.pdf
COGNEX (2021). What is the Difference between AI, Machine Learning, and Deep Learning for Industrial Automation Inspections?. Retrieved from https://www.cognex.com/blogs/deep-learning/ai-versus-deep-learning-versus-machine-learning-in-industrial-automation
Corradi, O.(2016). World Electricity Map. Retrieved from https://www.electricitymap.org/map
Cushman, W. H., & Rosenberg, D. J.(1991). Human Factors in Product Design. Netherlands: Elsevier.
Deshpande, A.(2016). A Beginner's Guide To Understanding Convolutional Neural Networks. Retrieved from https://adeshpande3.github.io/A-Beginner%27s-Guide-To-Understanding-Convolutional-Neural-Networks/
DMP (2018). A Closer Look at YOLOv3. Retrieved from
EllenMacArthurFoundation (2016). The New Plastics Economy: RETHINKING THE FUTURE OF PLASTICS. UK: Ellen MacArthur Foundation. Retrieved from https://www.ellenmacarthurfoundation.org/publications/the-new-plastics-economy-rethinking-the-future-of-plastics
EPA(2015). Global Greenhouse Gas Emissions Data. Retrieved from https://www.epa.gov/ghgemissions/global-greenhouse-gas-emissions-data
Eppinger, S., Ulrich, K.(2011). Product Design and Development. US: McGraw-Hill Education.
EU(2009). Sustainable Product Policy - Is 80% of environmental impact determined during the design phase. Retrieved from https://ec.europa.eu/jrc/en/research-topic/sustainable-product-policy#:~:text=It%20is%20estimated%20that%20over,throughout%20their%20entire%20life%20cycle
FIRA (2011). Benchmarking carbon footprints of furniture products. UK: FIRA international.
Freinkel, S. (2011). Plastic: A Toxic Love Story. US: Houghton Mifflin Harcourt.
Geyer, R. , Jenna R. Jambeck, & Law, K. L.(2017). Production, use, and fate of all plastics ever made. Science Advances, 3(7):e1700782.
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014). Rich feature hierarchies for accurate object detection and semantic segmentation. IEEE conference on computer vision and pattern recognition. 580-587.
Gosling, S. D., Craik, K. H., Martin, N. R., & Pryor, M. R. (2005). Material attributes of Personal Living Spaces. Home Cultures(2), 51-88.
Gosling, S. D., Craik, K. H., Martin, N. R., & Pryor, M. R. (2005). The Personal Living Space Cue Inventory:An Analysis and Evaluation. Environment and Behavior, 37(5), 683-705.
Graham, E. R., & Thompson, J. T.(2009). Deposit- and suspension-feeding sea cucumbers (Echinodermata) ingest plastic fragments . Journal of Experimental Marine Biology and Ecology, 368(1), 22-29.
Hannah Ritchie, & Roser, M. (2018). Plastic Pollution - Our World in Data . Retrieved from https://ourworldindata.org/plastic-pollution
Hao, K. (2019). Here are 10 ways AI could help fight climate change. Retrieved from https://www.technologyreview.com/2019/06/20/134864/ai-climate-change-machine-learning/
Hayes, D. (1990). Harnessing Market Forces to Protect the Earth. Science and Technology, 7(2), 46-51.
He, K., Gkioxari, G., Dollár, P., & Girshick, R. (2017). Mask R-CNN. IEEE international conference on computer vision, 2961-2969.
IDEO (2015). The Little Book of Design Research Ethics. US: IDEO.
IKEA (2019). Circular Product Design Guide: Guide to Navigating Through the Process of Designing Circular Products. Retrieved from https://preview.thenewsmarket.com/Previews/IKEA/DocumentAssets/512088_v2.pdf
ISO (2019). ISO 14064-2_2019(en), Greenhouse gases. Retrieved from https://www.iso.org/obp/ui#iso:std:iso:14064:-2:ed-2:v1:en
IUCN (2019). Review of plastic footprint methodologies: laying the foundation for the development of a standardised plastic footprint measurement tool. Switzerland: IUCN. DOI: https://doi.org/10.2305/IUCN.CH.2019.10.en
Kang, J., & Gwak, J. (2019). Ensemble of Instance Segmentation Models for Polyp Segmentation in Colonoscopy Images. IEEE Access, 7, 26440-26447.
Kanigel, R. (2007). Faux Real Genuine Leather and 200 Years of Inspired Fakes. US: University of Pennsylvania Press.
Kumar, R. (2014). Research Methodology: A Step-by-Step Guide for Beginners 4/e . US: SAGE Publications
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436-444.
Lynch, S. (2018). OpenLitterMap.com—Open Data on Plastic Pollution with Blockchain Rewards (Littercoin). Open Geospatial Data, Software and Standards, 3(1), p.6.
Malinowski, B. (1989). A Diary in the Strict Sense of the Term. US: Stanford University Press.
Malmodin, J., Moberg, Å., Lundén, D., Finnveden, G., & Lövehagen, N.(2010). Greenhouse Gas Emissions and Operational Electricity Use in the ICT and Entertainment & Media Sectors. Journal of Industrial Ecology, 14(5), 770-790.
McDonough, W., & Braungart, M.(2002). Cradle to Cradle: Remaking the Way We Make Things. US: North Point Press.
MinderooFoundation (2021). Plastic Waste Makers Index Revealing the source of the single-use plastics crisis. US: Minderoo Foundation. Retrieved from https://www.minderoo.org/plastic-waste-makers-index/findings/executive-summary/
Myers, K., & Secco, E. (2020). A Low-Cost Embedded Computer Vision System for the Classification of Recyclable Objects. Soft Computing Research Society and Congress on Intelligent Systems (CIS) 2020.
Neufeld, L. et al. (2016). The New Plastics Economy Rethinking the future of plastics. Switzerland: World Economic Forum.
OurWorldInData (2020). Sector by sector: where do global greenhouse gas emissions come from? Retrieved from https://ourworldindata.org/ghg-emissions-by-sector
Pack, A. T., Rose Phipps, E., Mattson, C. A., & Dahlin, E. C. (2019). Social Impact in Product Design, An Exploration of Current Industry Practices. Journal of Mechanical Design, 142(7): 071702.
PapersWithCode(2021). COCO test-dev Benchmark. Retrieved from https://paperswithcode.com/sota/object-detection-on-coco
Parker, L. (2017). A whopping 91% of plastic isn't recycled. Retrieved from https://www.nationalgeographic.com/
PDP(2010). Plastic Disclosure Project. Retrieved from https://www.plasticdisclosure.org/pdp
PlasticOceans (2021). The Basics On 7 Common Types of Plastic. Retrieved from https://plasticoceans.org/7-types-of-plastic/
Redmon, J., & Farhadi, A. (2018). Yolov3: An incremental improvement. Computer Vision and Pattern Recognition.
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. IEEE conference on computer vision and pattern recognition, 779-788.
Rees, W., & Wackernagel, M. (1996). Urban ecological footprints: Why cities cannot be sustainable—And why they are a key to sustainability. Environmental Impact Assessment Review, 16(4), 223-248.
Sandu, C. et al. (2020). Society Role in the Reduction of Plastic Pollution. The Handbook of Environmental Chemistry. pp.1-27. Berlin: Springer.
ScienceDirect (2013). Carbon Footprint definition. Retrieved from https://www.sciencedirect.com/topics/engineering/carbon-footprint
Tableau (2021). What is data cleaning?. Retrieved from https://www.tableau.com/learn/articles/what-is-data-cleaning
Tableau (2021).What is Tableau. Retrieved from https://www.tableau.com/zh-tw/why-tableau/what-is-tableau
Teuten, E. L., Saquing, J. M., Knappe, D. R., Barlaz, M. A., Jonsson, S., Bjorn, A. et al.(2009). Transport and release of chemicals from plastics to the environment and to wildlife. Philos Trans R Soc Lond B Biol Sci, 364(1526), 2027-2045.
ThoughtCo (2019). Most Common Plastics. Retrieved from https://www.thoughtco.com/most-common-plastics-820351
TOSHIBA(2008). TOSHIBA Explanation on the Calculation Method of the Factor. Retrieved from https://www.toshiba.co.jp/env/en/vision/factor2.htm
Toyota(2009). Toyota promotes Ecological Plastics in interiors. Retrieved from https://www.plasticstoday.com/toyota-promotes-ecological-plastics-interiors
UN (2015). Transforming Our World:The 2030 Agenda For Sustainable Development. US: United Nations.
UN (2019). Frontier Technology Quarterly: Frontier technologies for addressing plastic pollution. Retrieved from https://www.un.org/development/desa/dpad/publication/frontier-technology-quarterly-september-2019-frontier-technologies-for-addressing-plastic-pollution/
UN (2019). Frontier Technology Quarterly: Frontier technologies for addressing plastic pollution. Retrieved from https://www.un.org/development/desa/dpad/publication/frontier-technology-quarterly-september-2019-frontier-technologies-for-addressing-plastic-pollution/
UNEP (2014). Valuing plastics: The business case for measuring, managing and disclosing plastic use in the consumer goods industry. Kenya: UNEP.
UNEP (2021). UN-Beat plastic pollution. Kenya: UNEP. Retrieved from https://www.unep.org/interactive/beat-plastic-pollution/
UnileverGlobalCo. (2019). Reduce our virgin plastic packaging by 50% by 2025-Unilever announces ambitious new commitments for a waste-free world. Retrieved from https://www.unilever.com/news/press-releases/2019/unilever-announces-ambitious-new-commitments-for-a-waste-free-world.html
van Lieshout, C., van Oeveren, K., van Emmerik, T., & Postma, E. (2020). Automated River Plastic Monitoring Using Deep Learning and Cameras. Earth and Space Science, 7(8):e2019EA000960.
Vezzoli, C., & Manzini, E. (2007). Design per la sostenibilita ambientale. Italy: Zanichelli.
Wenzel, H., Hauschild, M. Z., & Alting, L. (1997). Environmental Assessment of Products: Methodology, Volume 1: Tools and Case Studies in Product Development. UK: Chapman and Hall.
Wu, S. (2013). A review on coarse warranty data and analysis. Reliability Engineering & System Safety, vol. 114, 1-11.

中文文獻
中華民國經濟部（2011）。產品綠色驗證檢索平台。 取自https://cogp.greentrade.org.tw/Certificate。
日經設計（2017）。非買不可! IKEA的設計。臺北市：天下文化。
王翊銘（2020）。從移動測繪資料自動化萃取交通號誌標誌以建置高精地圖。碩士論文。國立成功大學測量及空間資訊學系，台南市。
江俊毅（2015）。資訊通信產品之生態效率評估與綠色設計:以數位機上盒(Set-Top Box)為例。碩士論文。國立臺北大學自然資源與環境管理研究所在職專班，新北市。
吳亭慧（2019）。筆記型電腦滑鼠觸控板表面瑕疵檢測系統。碩士論文。國立臺北科技大學工業工程與管理系，台北市。
吳梅鳳（2011）。從產品碳足跡建構綠色產品管理系統。碩士論文。逢甲大學工業工程與系統管理學研究所，台中市。
呂宗翰（2016）。產品之環境足跡評估方法。碩士論文。國立成功大學機械工程學系，台南市。
宋同正（2019）。直達「循環設計圈」的效率設計準則 。取自https://cci.culture.tw/%E6%9C%AA%E4%BE%86%E7%94%9F%E5%AD%98%E4%B9%8B%E9%81%93%EF%BC%9A%E4%B8%80%E8%B5%B7%E3%80%8C%E5%BE%AA%E7%92%B0%E8%A8%AD%E8%A8%88%E3%80%8D%E5%90%A7%EF%BC%81/
李保寧（1999）。國外綠色文明的十二種趨勢－席捲全球的綠色浪潮。取自https://www.cnki.com.cn/Article/CJFDTotal-GJZW199808012.htm
杜瑞澤（2002）。產品永續設計-綠色設計理論與實務。臺北市：亞太圖書。
周念陵 （2011）。ISO14060系列溫室氣體相關標準發展現況。新北市：財團法人全國認證基金會。取自https://www.taftw.org.tw/wSite/ct?xItem=79&ctNode=89&mp=1
林大貴（2017）。TensorFlow + Keras 深度學習人工智慧實務應用。新北市：博碩文化。
林正祥（2010）。高性能工程塑膠次料回收產業技術創新之研究。碩士論文。國立中山大學高階經營碩士班，高雄市。
林滿盈（2010）。塑膠成型品設計與模具製作。新北市：全華圖書。
林澤宏（2020）。自行車零組件產品碳足跡盤查分析與資料庫建置。博士論文。國立勤益科技大學精密製造科技研究所，台中市。
張子成、邢繼綱（2008）。塑膠產品設計(修訂二版)。新北市：全華圖書。
張振亞、陳惠琳、董敏筑、鄭純宜（2018）。邁向台灣2030永續發展。台北市：財團法人資源循環台灣基金會。取自 https://circular-taiwan.org/learn/sdgs/
郭佩宜、王宏仁（2006）。田野的技藝：自我、研究與知識建構。高雄市：巨流圖書公司
陳永坤（2008）。台灣環境永續發展評估指標之建立與應用研究。博士論文。國立成功大學資源工程學系，台南市。
陳米蘭（2020）。國際氣候變遷局勢轉變：台灣的機會與挑戰。取自https://rsprc.ntu.edu.tw/zh-tw/m01-3/1520.html
勞動部（2020）。化學材料製造業。台灣：勞動部 取自https://www.mol.gov.tw/media/1664/ac05.pdf
曾福裕（2015）。電子產品碳足跡盤查及其綠色設計決策因子之研究。碩士論文。國立交通大學工學院永續環境科技學程，新竹市。
黃宜玉（2020）。海洋塑膠廢棄物污染防治—他國經驗與我國政策之比較。碩士論文。國立臺灣海洋大學海洋政策碩士學位學程，基隆市。
經濟部國貿局（2021）。搖籃到搖籃認證 Cradle to Cradle。取自https://cogp.greentrade.org.tw/Certificate/inside/117
詹景程（2014）。產品環境足跡盤查與組織溫室氣體減量管理－以半導體廠為例。碩士論文。國立臺北大學自然資源與環境管理研究所在職專班，新北市。
詹鈞婷（2019）。基於深度學習之心臟血管鈣化偵測系統。碩士論文。國立中央大學資訊工程學系，桃園縣。
劉海濤（2008）。人类学田野调查中的矛盾与困境。貴州民族研究，28（122）
劉翠溶（2013）。臺灣環境承載力之探測 — 初步結果。台北市：中央研究院。取自http://idv.sinica.edu.tw/ectjliu/%E5%8A%89%E7%BF%A0%E6%BA%B6%E5%AD%B8%E8%A1%93%E8%91%97%E4%BD%9C/W10-%E7%92%B0%E5%A2%83%E5%8F%B22pdf/2013%E6%89%BF%E8%BC%89%E5%8A%9B%E6%8E%A2%E6%B8%AC.pdf
鄭景隆（2011）。某一動態隨機存取記憶體的碳足跡盤查之研究。碩士論文。國立交通大學工學院永續環境科技學程，新竹市。
盧怡靜、呂穎彬（2014）。ISO 14040 生命週期評估的下一步。永續產業發展季刊，NO.66，頁29-35
環保署（2020）。行政院環境保護署推動產品碳足跡管理要點。 取自https://oaout.epa.gov.tw/law/LawContent.aspx?id=GL007755
謝志誠（2000）。ISO 14000文獻整理（台灣大學）。取自http://www.taiwan921.lib.ntu.edu.tw/mypdf/mf22.pdf