This paper investigates court congestion through a field study conducted at the Jerusalem District Court in Israel, aiming to reduce case processing time by adapting successful operational concepts to this unique environment. Using a modified difference-in-differences approach, the study suggests a notable 46.1% reduction in the duration of the treated part of the judicial process, demonstrating the efficacy of operational management tools in alleviating court congestion without compromising process quality or requiring additional resources.
{"title":"Alleviating Court Congestion: The Case of the Jerusalem District Court","authors":"Shany Azaria, B. Ronen, Noam Shamir","doi":"10.1287/inte.2023.0026","DOIUrl":"https://doi.org/10.1287/inte.2023.0026","url":null,"abstract":"This paper investigates court congestion through a field study conducted at the Jerusalem District Court in Israel, aiming to reduce case processing time by adapting successful operational concepts to this unique environment. Using a modified difference-in-differences approach, the study suggests a notable 46.1% reduction in the duration of the treated part of the judicial process, demonstrating the efficacy of operational management tools in alleviating court congestion without compromising process quality or requiring additional resources.","PeriodicalId":53206,"journal":{"name":"Informs Journal on Applied Analytics","volume":"14 3","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139009587","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yang Wang, Tong Wang, Xiaoqing Wang, Yuming Deng, Lei Cao
Improving fulfillment efficiency is critical for long-term sustainability of online grocery retailing. In this paper, we study reducing order fulfillment cost by order consolidation. Motivated by the observation that a significant percentage of buyers place multiple orders within a short time interval, we propose a scheme that attempts to consolidate such “multiorders” to reduce the number of parcels and hence, the shipping cost. At the same time, it cannot significantly disturb the existing order fulfillment process or undermine the customer service level. Successful execution of the scheme requires a prediction of multiorder probabilities and a control policy that selectively prioritizes order processing. For the prediction task, we formulate a binary classification problem and use machine-learning algorithms to predict in real time the probability of a multiorder. For the control task, our proposal is to hold arriving orders in a temporary order pool for potential consolidation and to determine the release timing by a dynamic program. The proposed solution is estimated to capture 92.8% of all the multiorders at the cost of holding the orders for about 20.3 minutes on average. This translates to more than 10 million U.S. dollars of order fulfillment cost saving annually. History: This paper was refereed.
{"title":"Data-Driven Order Fulfillment Consolidation for Online Grocery Retailing","authors":"Yang Wang, Tong Wang, Xiaoqing Wang, Yuming Deng, Lei Cao","doi":"10.1287/inte.2022.0068","DOIUrl":"https://doi.org/10.1287/inte.2022.0068","url":null,"abstract":"Improving fulfillment efficiency is critical for long-term sustainability of online grocery retailing. In this paper, we study reducing order fulfillment cost by order consolidation. Motivated by the observation that a significant percentage of buyers place multiple orders within a short time interval, we propose a scheme that attempts to consolidate such “multiorders” to reduce the number of parcels and hence, the shipping cost. At the same time, it cannot significantly disturb the existing order fulfillment process or undermine the customer service level. Successful execution of the scheme requires a prediction of multiorder probabilities and a control policy that selectively prioritizes order processing. For the prediction task, we formulate a binary classification problem and use machine-learning algorithms to predict in real time the probability of a multiorder. For the control task, our proposal is to hold arriving orders in a temporary order pool for potential consolidation and to determine the release timing by a dynamic program. The proposed solution is estimated to capture 92.8% of all the multiorders at the cost of holding the orders for about 20.3 minutes on average. This translates to more than 10 million U.S. dollars of order fulfillment cost saving annually. History: This paper was refereed.","PeriodicalId":53206,"journal":{"name":"Informs Journal on Applied Analytics","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136012847","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Inventory management is one of the most important components of Alibaba’s business. Traditionally, human buyers make replenishment decisions: although artificial intelligence (AI) algorithms make recommendations, human buyers can choose to ignore these recommendations and make their own decisions. The company has been exploring a new replenishment system in which algorithmic recommendations are final. The algorithms combine state-of-the-art deep reinforcement learning techniques with the framework of fictitious play. By learning the supplier’s behavior, we are able to address the important issues of lead time and fill rate on order quantity, which have been ignored in the extant literature of stochastic inventory control. We present evidence that our algorithms outperform human buyers in terms of reducing out-of-stock rates and inventory levels. More interestingly, we have seen additional benefits amid the pandemic. Over the last two years, cities in China partially and intermittently locked down to mitigate COVID-19 outbreaks. We have observed panic buying from human buyers during lockdowns, leading to the bullwhip effect. By contrast, panic buying and the bullwhip effect can be mitigated using our algorithms due to their ability to recognize changes in the supplier’s behavior during lockdowns. History: This paper has been accepted for the INFORMS Journal on Applied Analytics Special Issue—2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research.
库存管理是阿里巴巴业务最重要的组成部分之一。传统上，人类买家做出补货决策:尽管人工智能(AI)算法会提出建议，但人类买家可以选择忽略这些建议，自己做出决定。该公司一直在探索一种新的补货系统，其中算法推荐是最终的。该算法结合了最先进的深度强化学习技术和虚拟游戏框架。通过对供应商行为的学习，可以解决现有随机库存控制文献中忽略的交货时间和交货率对订单数量的重要影响。我们提供的证据表明，我们的算法在减少缺货率和库存水平方面优于人类买家。更有趣的是，我们在大流行期间看到了额外的好处。在过去两年中，中国的城市部分和间歇性封锁以缓解COVID-19疫情。我们观察到在封锁期间人类买家的恐慌性购买，导致牛鞭效应。相比之下，由于我们的算法能够识别封锁期间供应商行为的变化，因此可以缓解恐慌性购买和牛鞭效应。历史:本文已被INFORMS应用分析杂志特刊- 2022年Daniel H. Wagner高级分析和运筹学实践优秀奖所接受。
{"title":"AI vs. Human Buyers: A Study of Alibaba’s Inventory Replenishment System","authors":"Jiaxi Liu, Shuyi Lin, Linwei Xin, Yidong Zhang","doi":"10.1287/inte.2023.1160","DOIUrl":"https://doi.org/10.1287/inte.2023.1160","url":null,"abstract":"Inventory management is one of the most important components of Alibaba’s business. Traditionally, human buyers make replenishment decisions: although artificial intelligence (AI) algorithms make recommendations, human buyers can choose to ignore these recommendations and make their own decisions. The company has been exploring a new replenishment system in which algorithmic recommendations are final. The algorithms combine state-of-the-art deep reinforcement learning techniques with the framework of fictitious play. By learning the supplier’s behavior, we are able to address the important issues of lead time and fill rate on order quantity, which have been ignored in the extant literature of stochastic inventory control. We present evidence that our algorithms outperform human buyers in terms of reducing out-of-stock rates and inventory levels. More interestingly, we have seen additional benefits amid the pandemic. Over the last two years, cities in China partially and intermittently locked down to mitigate COVID-19 outbreaks. We have observed panic buying from human buyers during lockdowns, leading to the bullwhip effect. By contrast, panic buying and the bullwhip effect can be mitigated using our algorithms due to their ability to recognize changes in the supplier’s behavior during lockdowns. History: This paper has been accepted for the INFORMS Journal on Applied Analytics Special Issue—2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research.","PeriodicalId":53206,"journal":{"name":"Informs Journal on Applied Analytics","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134961607","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Theodore Papalexopoulos, James Alcorn, Dimitris Bertsimas, Rebecca Goff, Darren Stewart, Nikolaos Trichakis
In 2019, the United Network for Sharing (UNOS), which has been operating the Organ Procurement and Transplantation Network (OPTN) in the United States since 1984, was seeking to design a new national lung transplant allocation policy. The goal was to develop a point system that would prioritize candidates on the waiting list in a way that would yield more efficient and equitable outcomes. Our joint Massachusetts Institute of Technology (MIT)/UNOS team joined forces with the OPTN Lung Transplantation Committee in these policy design efforts. We discuss how our team applied a novel analytical framework, which was developed at MIT and utilizes optimization, regression, and simulation techniques, to illuminate salient trade-offs among outcomes and guide the choice of how to weigh different point attributes in the allocation formula. The committee selected for the allocation formula weights that were highlighted in the team’s analysis. The team’s proposal was implemented as the national lung allocation policy on March 9, 2023 across the United States. History: This paper has been accepted for the INFORMS Journal on Applied Analytics Special Issue—2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research.
2019年，自1984年以来一直在美国运营器官获取和移植网络(OPTN)的联合共享网络(UNOS)正在寻求设计一项新的国家肺移植分配政策。其目标是建立一个记分系统，将候补名单上的候选人按优先顺序排列，从而产生更有效和公平的结果。我们的麻省理工学院/UNOS联合团队与OPTN肺移植委员会在这些政策设计工作中通力合作。我们讨论了我们的团队如何应用一个新的分析框架，该框架是在麻省理工学院开发的，并利用优化、回归和模拟技术，来阐明结果之间的显著权衡，并指导如何在分配公式中权衡不同点属性的选择。委员会为分配公式选择了在团队分析中突出显示的权重。该团队的建议于2023年3月9日在美国全国范围内作为国家肺分配政策实施。历史:本文已被INFORMS应用分析杂志特刊- 2022年Daniel H. Wagner高级分析和运筹学实践优秀奖所接受。
{"title":"Applying Analytics to Design Lung Transplant Allocation Policy","authors":"Theodore Papalexopoulos, James Alcorn, Dimitris Bertsimas, Rebecca Goff, Darren Stewart, Nikolaos Trichakis","doi":"10.1287/inte.2023.0036","DOIUrl":"https://doi.org/10.1287/inte.2023.0036","url":null,"abstract":"In 2019, the United Network for Sharing (UNOS), which has been operating the Organ Procurement and Transplantation Network (OPTN) in the United States since 1984, was seeking to design a new national lung transplant allocation policy. The goal was to develop a point system that would prioritize candidates on the waiting list in a way that would yield more efficient and equitable outcomes. Our joint Massachusetts Institute of Technology (MIT)/UNOS team joined forces with the OPTN Lung Transplantation Committee in these policy design efforts. We discuss how our team applied a novel analytical framework, which was developed at MIT and utilizes optimization, regression, and simulation techniques, to illuminate salient trade-offs among outcomes and guide the choice of how to weigh different point attributes in the allocation formula. The committee selected for the allocation formula weights that were highlighted in the team’s analysis. The team’s proposal was implemented as the national lung allocation policy on March 9, 2023 across the United States. History: This paper has been accepted for the INFORMS Journal on Applied Analytics Special Issue—2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research.","PeriodicalId":53206,"journal":{"name":"Informs Journal on Applied Analytics","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134961601","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-01DOI: 10.1287/inte.2023.intro.v53.n5
Margret V. Bjarnadottir, Lawrence D. Stone
The judges for the 2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research selected the four finalist papers featured in this special issue of the INFORMS Journal on Applied Analytics (IJAA). The prestigious Wagner Prize—awarded for achievement in implemented operations research, management science, and advanced analytics—emphasizes the quality and originality of mathematical models along with clarity of written and oral exposition. This year’s winning application describes the design and deployment of a generalized synthetic control, a powerful and innovative statistical method for identifying, in a noisy environment, retailing innovations that produce a small percentage improvement in a large volume of sales for Anheuser Busch Inbev. The remaining three papers describe an inverse control approach to allocating lung transplants that best meets targeted outcomes and has been implemented as the national lung allocation policy on March 9, 2023, across the United States; a human-centric, optimized parcel delivery system developed for Deutsche Post that saves money while meeting constraints learned dynamically from driver behavior; and an AI-based system developed for Alibaba that learns supplier behavior to improve replenishment ordering and inventory control. Supplemental Material: Full presentation videos with slides are available in the INFORMS Video Library at https://www.informs.org/Resource-Center/Video-Library and as electronic companions to the INFORMS Journal on Applied Analytics articles.
2022年Daniel H. Wagner高级分析和运筹学实践卓越奖的评委选出了本期INFORMS应用分析杂志(IJAA)特刊上的四篇入围论文。著名的瓦格纳奖(Wagner prize)——授予在实施运筹学、管理科学和高级分析方面取得成就的人——强调数学模型的质量和原创性，以及书面和口头阐述的清晰度。今年的获奖申请描述了一种广义综合控制的设计和部署，这是一种强大而创新的统计方法，用于在嘈杂的环境中识别零售创新，这些创新可以为安海斯布希英博(Anheuser Busch Inbev)的大量销售额带来很小的百分比提高。其余三篇论文描述了分配肺移植的逆控制方法，该方法最能满足目标结果，并已于2023年3月9日在全美范围内作为国家肺分配政策实施;为德国邮政(Deutsche Post)开发的以人为本的优化包裹递送系统，既节省了资金，又满足了从驾驶员行为中动态学习到的约束;以及为阿里巴巴开发的基于人工智能的系统，该系统可以学习供应商的行为，以改善补货订单和库存控制。补充材料:完整的演示视频和幻灯片可以在INFORMS视频库中获得，网址为https://www.informs.org/Resource-Center/Video-Library，也可以作为INFORMS应用分析期刊文章的电子同伴。
{"title":"Introduction: 2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research","authors":"Margret V. Bjarnadottir, Lawrence D. Stone","doi":"10.1287/inte.2023.intro.v53.n5","DOIUrl":"https://doi.org/10.1287/inte.2023.intro.v53.n5","url":null,"abstract":"The judges for the 2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research selected the four finalist papers featured in this special issue of the INFORMS Journal on Applied Analytics (IJAA). The prestigious Wagner Prize—awarded for achievement in implemented operations research, management science, and advanced analytics—emphasizes the quality and originality of mathematical models along with clarity of written and oral exposition. This year’s winning application describes the design and deployment of a generalized synthetic control, a powerful and innovative statistical method for identifying, in a noisy environment, retailing innovations that produce a small percentage improvement in a large volume of sales for Anheuser Busch Inbev. The remaining three papers describe an inverse control approach to allocating lung transplants that best meets targeted outcomes and has been implemented as the national lung allocation policy on March 9, 2023, across the United States; a human-centric, optimized parcel delivery system developed for Deutsche Post that saves money while meeting constraints learned dynamically from driver behavior; and an AI-based system developed for Alibaba that learns supplier behavior to improve replenishment ordering and inventory control. Supplemental Material: Full presentation videos with slides are available in the INFORMS Video Library at https://www.informs.org/Resource-Center/Video-Library and as electronic companions to the INFORMS Journal on Applied Analytics articles.","PeriodicalId":53206,"journal":{"name":"Informs Journal on Applied Analytics","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134961606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Uğur Arıkan, Thorsten Kranz, Baris Cem Sal, Severin Schmitt, Jonas Witt
Features such as estimated delivery time windows and live tracking of shipments play a key role in improving the customer experience in last-mile delivery. The building blocks for enabling these features are reliable knowledge about the expected order of deliveries in a tour and precise delivery time window predictions. For Deutsche Post’s parcel delivery service in Germany, we developed a courier-centric routing algorithm and a corresponding state-of-the-art machine learning model for delivery time window predictions. The routing algorithm combines operations research with statistics and machine learning to implicitly gather and use the tacit knowledge of our experienced couriers within the tour generation. This is achieved by deducing and selecting appropriate precedence constraints from historical delivery data. This novel combination of optimization with data-driven constraints enabled us to provide custom routes to the individual couriers. It proved to be a main driver allowing us to provide accurate delivery time window predictions and live tracking of shipments. Our solution is used by Deutsche Post to plan the daily routes of couriers to the approximately 13,000 parcel delivery districts in Germany as well as to provide live tracking and estimated delivery time windows for 1.6 million parcels each day. History: This paper has been accepted for the INFORMS Journal on Applied Analytics Special Issue—2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research.
预计交货时间窗口和实时货物跟踪等功能在改善最后一英里交货的客户体验方面发挥着关键作用。支持这些特性的构建块是关于旅行中预期交付顺序的可靠知识和精确的交付时间窗口预测。对于德国邮政的包裹递送服务，我们开发了一个以快递员为中心的路由算法和一个相应的最先进的机器学习模型，用于投递时间窗口预测。路由算法将运筹学与统计学和机器学习相结合，隐式地收集和使用我们在旅行一代中经验丰富的快递员的隐性知识。这是通过从历史交付数据中推断和选择适当的优先约束来实现的。这种优化与数据驱动约束的新颖结合使我们能够为个人快递员提供定制路线。事实证明，它是一个主要的驱动因素，使我们能够提供准确的交货时间窗口预测和实时跟踪货物。德国邮政使用我们的解决方案来规划快递员前往德国约13,000个包裹投递区的每日路线，并每天为160万个包裹提供实时跟踪和估计投递时间窗口。历史:本文已被INFORMS应用分析杂志特刊- 2022年Daniel H. Wagner高级分析和运筹学实践优秀奖所接受。
{"title":"Human-Centric Parcel Delivery at Deutsche Post with Operations Research and Machine Learning","authors":"Uğur Arıkan, Thorsten Kranz, Baris Cem Sal, Severin Schmitt, Jonas Witt","doi":"10.1287/inte.2023.0031","DOIUrl":"https://doi.org/10.1287/inte.2023.0031","url":null,"abstract":"Features such as estimated delivery time windows and live tracking of shipments play a key role in improving the customer experience in last-mile delivery. The building blocks for enabling these features are reliable knowledge about the expected order of deliveries in a tour and precise delivery time window predictions. For Deutsche Post’s parcel delivery service in Germany, we developed a courier-centric routing algorithm and a corresponding state-of-the-art machine learning model for delivery time window predictions. The routing algorithm combines operations research with statistics and machine learning to implicitly gather and use the tacit knowledge of our experienced couriers within the tour generation. This is achieved by deducing and selecting appropriate precedence constraints from historical delivery data. This novel combination of optimization with data-driven constraints enabled us to provide custom routes to the individual couriers. It proved to be a main driver allowing us to provide accurate delivery time window predictions and live tracking of shipments. Our solution is used by Deutsche Post to plan the daily routes of couriers to the approximately 13,000 parcel delivery districts in Germany as well as to provide live tracking and estimated delivery time windows for 1.6 million parcels each day. History: This paper has been accepted for the INFORMS Journal on Applied Analytics Special Issue—2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research.","PeriodicalId":53206,"journal":{"name":"Informs Journal on Applied Analytics","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134962200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Luis Costa, Vivek F. Farias, Patricio Foncea, Jingyuan (Donna) Gan, Ayush Garg, Ivo Rosa Montenegro, Kumarjit Pathak, Tianyi Peng, Dusan Popovic
We describe a novel optimization-based approach—generalized synthetic control (GSC)—in which we learn from experiments conducted in a physical retail environment. GSC solves a long-standing problem of learning from experiments conducted in this environment when treatment effects are small, the environment is extremely noisy and nonstationary, and interference and adherence problems are commonplace. The utilization of GSC has demonstrated a remarkable increase in statistical power, approximately one hundredfold (100×) higher than conventional inferential methods. This innovative approach forms the basis of TestOps, a pioneering large-scale experimentation platform designed specifically for physical retailers. TestOps was developed and has been broadly implemented as part of a collaboration between Anheuser Busch Inbev (ABI) and a team of operations researchers and data engineers from the Massachusetts Institute of Technology. TestOps currently runs physical experiments impacting approximately 135 million USD in revenue every month and routinely identifies innovations that result in a 1%–2% increase in sales volume. The vast majority of these innovations would have remained unidentified had we not developed our novel approach to inference. Prior to our implementation, statistically significant conclusions could be drawn on only ∼6% of all experiments, a fraction that has now increased by 10-fold. Given its success, TestOps is being rolled out globally at ABI, driving significant revenue growth and enabling the extraction of valuable insights from large-scale physical experiments. History: This paper has been accepted for the INFORMS Journal on Applied Analytics Special Issue—2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research.
我们描述了一种新的基于优化的方法-广义综合控制(GSC) -我们从物理零售环境中进行的实验中学习。GSC解决了一个长期存在的问题，即从在这种环境下进行的实验中学习，在这种环境下，治疗效果很小，环境非常嘈杂和非平稳，干扰和粘附问题很常见。GSC的使用在统计能力上有了显著的提高，大约比传统的推理方法高100倍。这种创新的方法构成了TestOps的基础，TestOps是一个专门为实体零售商设计的开创性的大规模实验平台。TestOps是Anheuser Busch Inbev (ABI)与麻省理工学院的运营研究人员和数据工程师团队合作开发并广泛实施的。TestOps目前进行的物理实验每月影响约1.35亿美元的收入，并定期识别导致销售额增加1%-2%的创新。如果我们没有开发出新的推理方法，这些创新中的绝大多数都不会被发现。在我们实施之前，只有6%的实验可以得出具有统计学意义的结论，这一比例现在增加了10倍。鉴于它的成功，TestOps正在ABI的全球范围内推广，推动了显著的收入增长，并使从大规模物理实验中提取有价值的见解成为可能。历史:本文已被INFORMS应用分析杂志特刊- 2022年Daniel H. Wagner高级分析和运筹学实践优秀奖所接受。
{"title":"Generalized Synthetic Control for TestOps at ABI: Models, Algorithms, and Infrastructure","authors":"Luis Costa, Vivek F. Farias, Patricio Foncea, Jingyuan (Donna) Gan, Ayush Garg, Ivo Rosa Montenegro, Kumarjit Pathak, Tianyi Peng, Dusan Popovic","doi":"10.1287/inte.2023.0028","DOIUrl":"https://doi.org/10.1287/inte.2023.0028","url":null,"abstract":"We describe a novel optimization-based approach—generalized synthetic control (GSC)—in which we learn from experiments conducted in a physical retail environment. GSC solves a long-standing problem of learning from experiments conducted in this environment when treatment effects are small, the environment is extremely noisy and nonstationary, and interference and adherence problems are commonplace. The utilization of GSC has demonstrated a remarkable increase in statistical power, approximately one hundredfold (100×) higher than conventional inferential methods. This innovative approach forms the basis of TestOps, a pioneering large-scale experimentation platform designed specifically for physical retailers. TestOps was developed and has been broadly implemented as part of a collaboration between Anheuser Busch Inbev (ABI) and a team of operations researchers and data engineers from the Massachusetts Institute of Technology. TestOps currently runs physical experiments impacting approximately 135 million USD in revenue every month and routinely identifies innovations that result in a 1%–2% increase in sales volume. The vast majority of these innovations would have remained unidentified had we not developed our novel approach to inference. Prior to our implementation, statistically significant conclusions could be drawn on only ∼6% of all experiments, a fraction that has now increased by 10-fold. Given its success, TestOps is being rolled out globally at ABI, driving significant revenue growth and enabling the extraction of valuable insights from large-scale physical experiments. History: This paper has been accepted for the INFORMS Journal on Applied Analytics Special Issue—2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research.","PeriodicalId":53206,"journal":{"name":"Informs Journal on Applied Analytics","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134962202","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kiera W. Dobbs, Rahul Swamy, D. King, Ian G. Ludden, S. Jacobson
Every 10 years, U.S. states redraw their congressional and state legislative district plans. This process decides the political landscape for the subsequent 10 years. Prior to the 2021 redistricting cycle, Missouri enacted new criteria for state legislative districts. The Missouri League of Women Voters (LWV-MO) contacted the authors to analyze the potential impact of these new criteria on the map-drawing process. We apply recombination (a spanning tree method) within a local search optimization framework to analyze the interplay between political geography, constitutional requirements, and political fairness in Missouri. We use this framework to produce district plans that satisfy the new criteria and prioritize different aspects of fairness. The results, quantified by several measures of fairness, reveal an inherent Republican advantage in Missouri because of the state’s political geography and constitutional requirements. We conclude that Missouri’s political geography and constitutional requirements prevent the optimization framework from substantially improving political fairness in state legislative plans. In contrast, the framework can substantially improve political fairness in Missouri congressional plans, which are not subject to the new requirements. The LWV-MO used this work to advocate for fairness and transparency in their testimonies for the Missouri redistricting commission’s public hearings. History: This paper was refereed. Funding: This material is based upon work supported by the National Science Foundation Graduate Research Fellowship Program [Grant DGE-1746047]. S. H. Jacobson was supported by the Air Force Office of Scientific Research [Grant FA9550-19-1-0106]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/inte.2022.0037 .
每隔10年，美国各州都会重新划分国会选区和州立法区。这一进程决定了今后10年的政治格局。在2021年重新划分选区之前，密苏里州为州立法区制定了新的标准。密苏里州妇女选民联盟(LWV-MO)联系了作者，分析了这些新标准对地图绘制过程的潜在影响。我们在本地搜索优化框架内应用重组(生成树方法)来分析密苏里州的政治地理、宪法要求和政治公平之间的相互作用。我们使用这个框架来制定符合新标准的地区计划，并优先考虑公平性的不同方面。通过若干公平指标量化的结果显示，由于该州的政治地理和宪法要求，共和党在密苏里州具有固有的优势。我们的结论是，密苏里州的政治地理和宪法要求阻碍了优化框架在州立法计划中大幅提高政治公平性。相比之下，该框架可以大大提高密苏里州国会计划的政治公平性，这些计划不受新要求的约束。LWV-MO利用这项工作在密苏里州选区重新划分委员会的公开听证会上倡导公平和透明的证词。历史:本文被审稿。基金资助:本材料基于国家科学基金研究生研究奖学金计划[Grant DGE-1746047]支持的工作。S. H. Jacobson项目得到了美国空军科学研究办公室的支持[Grant FA9550-19-1-0106]。补充材料:在线附录可在https://doi.org/10.1287/inte.2022.0037上获得。
{"title":"An Optimization Case Study in Analyzing Missouri Redistricting","authors":"Kiera W. Dobbs, Rahul Swamy, D. King, Ian G. Ludden, S. Jacobson","doi":"10.1287/inte.2022.0037","DOIUrl":"https://doi.org/10.1287/inte.2022.0037","url":null,"abstract":"Every 10 years, U.S. states redraw their congressional and state legislative district plans. This process decides the political landscape for the subsequent 10 years. Prior to the 2021 redistricting cycle, Missouri enacted new criteria for state legislative districts. The Missouri League of Women Voters (LWV-MO) contacted the authors to analyze the potential impact of these new criteria on the map-drawing process. We apply recombination (a spanning tree method) within a local search optimization framework to analyze the interplay between political geography, constitutional requirements, and political fairness in Missouri. We use this framework to produce district plans that satisfy the new criteria and prioritize different aspects of fairness. The results, quantified by several measures of fairness, reveal an inherent Republican advantage in Missouri because of the state’s political geography and constitutional requirements. We conclude that Missouri’s political geography and constitutional requirements prevent the optimization framework from substantially improving political fairness in state legislative plans. In contrast, the framework can substantially improve political fairness in Missouri congressional plans, which are not subject to the new requirements. The LWV-MO used this work to advocate for fairness and transparency in their testimonies for the Missouri redistricting commission’s public hearings. History: This paper was refereed. Funding: This material is based upon work supported by the National Science Foundation Graduate Research Fellowship Program [Grant DGE-1746047]. S. H. Jacobson was supported by the Air Force Office of Scientific Research [Grant FA9550-19-1-0106]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/inte.2022.0037 .","PeriodicalId":53206,"journal":{"name":"Informs Journal on Applied Analytics","volume":"27 1","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73263646","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Robert M. Curry, Joseph Foraker, G. Lazzaro, David M Ruth
The U.S. Naval Academy is composed of 30 companies of students. Each student has a merit score, and each company has an average merit score. Leadership desires to minimize the deviation in average merit scores by splitting each company into first-year and upper-class groups and reassigning first-year groups to new upper-class groups. We perform this reassignment using greedy and optimal approaches. The standard deviation of average merit scores is reduced by more than half. History: This paper was refereed. Funding: This work was supported by the Office of Naval Research Global [Grant N0001421WX01983].
{"title":"Practice Summary: Optimal Student Group Reassignment at U.S. Naval Academy","authors":"Robert M. Curry, Joseph Foraker, G. Lazzaro, David M Ruth","doi":"10.1287/inte.2022.0055","DOIUrl":"https://doi.org/10.1287/inte.2022.0055","url":null,"abstract":"The U.S. Naval Academy is composed of 30 companies of students. Each student has a merit score, and each company has an average merit score. Leadership desires to minimize the deviation in average merit scores by splitting each company into first-year and upper-class groups and reassigning first-year groups to new upper-class groups. We perform this reassignment using greedy and optimal approaches. The standard deviation of average merit scores is reduced by more than half. History: This paper was refereed. Funding: This work was supported by the Office of Naval Research Global [Grant N0001421WX01983].","PeriodicalId":53206,"journal":{"name":"Informs Journal on Applied Analytics","volume":"29 1","pages":""},"PeriodicalIF":1.4,"publicationDate":"2023-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76289710","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：