[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124009-en":3,"doc-seo-124009-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},124009,8796095461564,"Liam","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Management Decisions in Manufacturing using Causal Machine Learning - To Rework, or not to Rework?","Data-driven decision model for rework in manufacturing systems that may optionally repair items within a multistage, lot-based process. Rework choice is made using the intermediate lot and system state, while product inspection and yield assessment are delayed until completion. Repair actions are uniform across the lot, creating a trade-off between potential yield gains for some items and degradation for others. A causal machine learning framework estimates yield improvement via double/debiased machine learning and derives optimal rework policies from data, validated on real opto-electronic semiconductor manufacturing with 2–3% yield improvement.","Management Decisions in Manufacturing using Causal Machine Learning – To Rework, or not to Rework?  \narXiv :2406 . 11308v1 [ cs .LG] 17 Jun 2024  \nPhilipp Schwarz∗ University of Hamburg  \nams Osram [philipp. schwarz@ams-osram. com](philipp. schwarz@ams-osram. com)  \nOliver Schacht∗  \nUniversity of Hamburg [oliver. schacht@uni-hamburg. de](oliver. schacht@uni-hamburg. de)  \nSven Klaassen∗  \nUniversity of Hamburg Economic AI  \nDaniel Grünbaumams Osram  \nSebastian Imhofams Osram  \nMartin Spindler  \nUniversity of Hamburg  \nEconomic AI  \nAbstract  \nIn this paper, we present a data-driven model for estimating optimal rework policies in manufacturing systems. We consider a single production stage within a multistage, lot-based system that allows for optional rework steps. While the rework decision depends on an intermediate state of the lot and system, the final product inspection, and thus the assessment of the actual yield, is delayed until production is complete. Repair steps are applied uniformly to the lot, potentially improving some of the individual items while degrading others. The challenge is thus to balance potential yield improvement with therework costs incurred. Given the inherently causal nature of this decision problem, we propose a causal model to estimate yield improvement. We apply methods from causal machine learning, in particular double/debiased machine learning (DML) techniques, to estimate conditional treatment effects from data and derive policies for rework decisions. We validate our decision model using real-world data from opto-electronic semiconductor manufacturing, achieving a yield improvement of 2 − 3% during the color-conversion process of white light-emitting diodes (LEDs) .  \nKeywords: Causal Machine Learning, Heterogeneous Treatment Effects, Double Machine Learning, Policy Learning, Rework, Multistage Manufacturing, Production Optimization, Phosphor-converted White LEDs  \n1 Introduction  \nCost considerations and the increasing importance of sustainability require efficient manufacturing systems that produce high-quality items. However, in complex value chains, such as semiconductor manufacturing, imperfect processes can lead to products that fail to meet the required quality targets. Several strategies have  \nbeen proposed to avoid, reduce the number of, or cope with defective products with the goal of minimizing ∗ Equal Contribution.  \ncosts. While preventing defects is preferable to mitigating them, doing so usually assumes an in-depth knowledge of the processes involved (see Powell et al. [2022]) . In cases where this assumption does not hold, defect compensation techniques, such as inline rework, seem more desirable than discard strategies. We argue that in the case of rework, a well-informed economic decision should be made regarding whether to repair a specific item. When dealing with an imperfect rework process in lot-based production systems, the trade-off between the additional costs of and the savings from the repair step is even more apparent. Indeed, depending on their state, some items in the production lot might benefit from the repair while others deteriorate. Thus, a decision model is needed that minimizes the yield loss incurred through the optional repair steps.  \nIn the field of production planning and inventory control, rework has been intensively studied, with a primary focus on improving the logistical decision-making associated with the rework process (e.g., Wein [1992], So and Tang [1995], Alsawafy and Selim [2022]) . Although logistical process properties such as lot size, production, and inspection cycles affect the number of defectives (e.g., Lee and Rosenblatt [1987]), the actual impact of rework on the product has not been a major concern in this branch of research. Except for Colledani and Angius [2020], the product state has usually been ignored, leading to simplifications that do not allow for individual repair decisions.  \nStatistical quality control and improvement","cbCaikBKLTmkT19o","https://ap.wps.com/l/cbCaikBKLTmkT19o","pdf",1331104,1,30,"English","en",105,"# Abstract\n# Introduction\n## Problem motivation: cost, quality, and sustainability\n## Limitations of existing rework and quality-control research\n## Relation to zero defect manufacturing and inline rework\n## Causal reasoning background","[{\"question\":\"How is the rework decision modeled in the manufacturing system?\",\"answer\":\"The model estimates optimal rework policies using the intermediate lot and system state, together with the delayed final product inspection that determines realized yield.\"},{\"question\":\"Why can rework improve some items while degrading others?\",\"answer\":\"In lot-based production, repair steps are applied uniformly, so depending on each item's intermediate state, the same action can lead to heterogeneous outcomes.\"},{\"question\":\"What causal machine learning method is used to estimate the effect of rework?\",\"answer\":\"The study uses double/debiased machine learning (DML) to estimate conditional treatment effects from data and then derive policies for rework decisions.\"}]","Management Decisions in Manufacturing using Causal Machine Learning - To Rework, or not to Rework? | PDF",1785819802,76,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"management-decisions-in-manufacturing-using-causal-machine-learning-to-rework-or-not-to-rework","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/management-decisions-in-manufacturing-using-causal-machine-learning-to-rework-or-not-to-rework/124009/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"How is the rework decision modeled in the manufacturing system?","Question",{"text":75,"@type":76},"The model estimates optimal rework policies using the intermediate lot and system state, together with the delayed final product inspection that determines realized yield.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why can rework improve some items while degrading others?",{"text":80,"@type":76},"In lot-based production, repair steps are applied uniformly, so depending on each item's intermediate state, the same action can lead to heterogeneous outcomes.",{"name":82,"@type":73,"acceptedAnswer":83},"What causal machine learning method is used to estimate the effect of rework?",{"text":84,"@type":76},"The study uses double/debiased machine learning (DML) to estimate conditional treatment effects from data and then derive policies for rework decisions.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":21,"slug":121},"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]