abstract: Supply chains are increasingly complex as companies branch out into newer products and markets. In many cases, multiple products with moderate differences in performance and price compete for the same unit of demand. Simultaneous occurrences of multiple scenarios (competitive, disruptive, regulatory, economic, etc.), coupled with business decisions (pricing, product introduction, etc.) can drastically change demand structures within a short period of time. Furthermore, product obsolescence and cannibalization are real concerns due to short product life cycles. Analytical tools that can handle this complexity are important to quantify the impact of business scenarios/decisions on supply chain performance. Traditional analysis methods struggle in this environment of large, complex datasets with hundreds of features becoming the norm in supply chains. We present an empirical analysis framework termed Scenario Trees that provides a novel representation for impulse and delayed scenario events and a direction for modeling multivariate constrained responses. Amongst potential learners, supervised learners and feature extraction strategies based on tree-based ensembles are employed to extract the most impactful scenarios and predict their outcome on metrics at different product hierarchies. These models are able to provide accurate predictions in modeling environments characterized by incomplete datasets due to product substitution, missing values, outliers, redundant features, mixed variables and nonlinear interaction effects. Graphical model summaries are generated to aid model understanding. Models in complex environments benefit from feature selection methods that extract non-redundant feature subsets from the data. Additional model simplification can be achieved by extracting specific levels/values that contribute to variable importance. We propose and evaluate new analytical methods to address this problem of feature value selection and study their comparative performance using simulated datasets. We show that supply chain surveillance can be structured as a feature value selection problem. For situations such as new product introduction, a bottom-up approach to scenario analysis is designed using an agent-based simulation and data mining framework. This simulation engine envelopes utility theory, discrete choice models and diffusion theory and acts as a test bed for enacting different business scenarios. We demonstrate the use of machine learning algorithms to analyze scenarios and generate graphical summaries to aid decision making. / Dissertation/Thesis / Ph.D. Industrial Engineering 2012

The short life cycles and highly variable demand in the fast fashion market causes various challenges in a retailer’s supply chain management processes. The essential task at hand is to provide the right product, at the right time, and at the right place. Due to this inherent difficulty, the bullwhip effect is a major issue in the fashion supply chain. To enhance customer satisfaction and increase the alignment between the supply and market place demand, companies have been pushed towards exploiting big data, supply chain analytics and AI techniques for better business decision making. One such critical but intrinsically complex decision is the development of a future apparel assortment; in particular defining its optimal breadth and depth. This thesis investigates how such AI techniques can be applied to develop a new assortment aligned with the future customer demands- and choice behavior. The research was conducted through firstly performing a qualitative case study at a fast fashion retailer. This explored the critical business decisions in the supply chain lacking AI support. The findings, revealing the assortment planning process as one such critical area, guided the second part of the thesis: a systematic literature review exploring the AI techniques used in this process in the retail - and fashion industry. An appropriate framework of planning a static apparel assortment in the fast fashion industry was developed and used as a guide throughout the study. The thesis discovered that there exists significant research in the field of applying AI techniques to generate and integrate knowledge about consumer demand- and choice behavior in the planning process of a future assortment. The main components to consider in this procedure is a) fashion forecasting, b) forecasting midterm demand, and c) forecasting product selection, incorporating the effects of substitution and complementarity at all times. This is believed to increase the alignment between supply and the marketplace demand, consequently reducing the bullwhip effect. The critical area for future research is how the discovered models are to be integrated in one singlemodel. Namely, simultaneously utilizing consumer choice behavior models and fashion forecasting to predict future demand of new items. Thus, the risk of suboptimization may be mitigated. / Modeindustrins korta produktlivscykler och högt varierande efterfrågan efter rådande trender skapar stora utmaningar i försörjningskedjan hos företag i branschen. Det essentiella målet för företagen är att tillhandahålla rätt produkt, vid rätt tidpunkt och på rätt plats. De komplexa karaktärsdragen i modeindustrin, där bland den fluktuerande efterfrågan, har gjort bullwhipeffekten till en stor utmaning i branschen. För att öka kundnöjdhet och anpassningen mellan marknadens utbud och efterfrågan har företag drivits mot utnyttjandet av big data i avsikt att förbättra kritisk affärsbeslutsfattning genom användandet av analytics och AI. Ett kritiskt ochkomplext beslut är utvecklingen av ett nytt produktsortiment, där definieringen av sortimentetsbredd och djup är särskilt viktigt. Denna uppsats undersöker hur AI-modeller kan tillämpas för att hjälpa företag inom modeindustrin i utvecklingen av nya sortiment anpassade efter kundens beräknade efterfrågan och val. Detta arbete inleddes med utförandet av en kvalitativ fallstudie hos en stor aktör verksam inom modeindustrin. Detta gjordes för att identifiera kritiska affärsbeslut i företagets försörjningskedja som saknade AI-stöd. Resultatet påvisade att sortimentsplanering var ett sådant kritiskt beslutsområde. Följaktligen utfördes en systematisk litteraturstudie i andra delen av arbetet i syfte att undersöka AI-modeller som appliceras i sortimentsplanerings-processen i såväl detaljhandeln som modebranschen. För att konceptualisera processen av att planera ett statiskt produktsortiment utvecklades ett ramverk som användes som en guide under hela arbetet. Studien visade att det finns betydande forskning inom tillämpningen av AI-modeller i syfte att planera ett optimalt sortiment efter konsumenternas efterfrågan. De huvudsakliga faktorerna att överväga innefattar prognostiseringen av efterfrågan, trender samt substitution- och komplementeffekter. Ett kritiskt område för framtida forskning är hur de upptäckta modellerna ska integreras i en enda modell som inkluderar dessa faktorer i ett tidigt såväl som sent skede av planeringen. Det som eftersträvas i en integrerad modell är att mildra risken av suboptimering som identifierats i nuvarande litteraturs angreppssätt.

<p>Developing resiliency in supply chains became vital in the recent years due to global diversification and vulnerability to risks. Firms need to identify, evaluate, and mitigate risks in supply chain to maintain continuity and create competitive advantage. Although the problem of supply chain disruptions has existed for a long time, less attention has been given by researchers in exploring the adoption of advanced technologies to build resilient supply chains. This study explored the potential of mitigating supply chain disruptions with the use of Industry 4.0 technologies such as Internet of Things (IoT) and Supply chain data analytics platform which develops digital twin environment for supply chains. </p> <p><br></p> <p>This research gathered expert’s opinion on the resilience capabilities developed in supply chain by digital twin adoption, stages and practices involved in digital twin assimilation through Delphi survey with subject matter experts and supply chain practitioners. Semi-structured interviews were conducted with participants to attain deep understanding on the resilience capabilities gained by digital twin and stages in digital twin adoption. Comparison of the results from Delphi survey and interviews was carried out to synthesize the results to yield a comprehensive understanding of resilience capabilities gained through digital twin and adoption stages of supply chain digital twin. This research has conducted interviews with 21 subject matter experts and completed three rounds of Delphi survey (with participants n = 15, 11, 11 in three rounds respectively) to develop a framework for digital twin adoption to enhance supply chain resilience. </p> <p><br></p> <p>This research determined that digital twin develops real-time monitoring and sensing capabilities, planning and decision support system, and automating decisions and action execution capabilities in supply chain. In addition, digital twin positively impacts resilience elements such as agility, supply chain reconfiguration, robustness, and collaboration in supply chain, which improves the supply chain performance. The results from this study were utilized to develop a framework for enabling supply chain resilience through digital twin. The framework included antecedents, consequences, and various moderators that impact digital twin adoption and diffusion in supply chains. Finally, this research developed a five-stage roadmap for adopting digital twin capabilities in supply chain. </p>

Today’s dynamic and increasingly competitive market had expanded complexities for global businesses pressuring companies to start leveraging on Big Data solutions in order to sustain the global competitions by becoming more data-driven in managing their supply chains.The main purpose of this study is twofold, 1) to explore the implications of applying analytics designing supply chain networks, 2) to investigate the link between operational and strategic management levels when making strategic decisions using Analytics.Qualitative methods have been applied for this study to gain a greater understanding of the Supply Chain Analytics phenomenon. An inductive approach in form of interviews, was performed in order to gain new empirical data. Fifteen semi-structured interviews were conducted with professional individuals who hold managerial roles such as project managers, consultants, and end-users within the fields of Supply Chain Management and Big Data Analytics. The received empirical information was later analyzed using the thematic analysis method.The main findings in this thesis relatively contradicts with previous studies and existing literature in terms of connotations, definitions and applications of the three main types of Analytics. Furthermore, the findings present new approaches and perspectives that advanced analytics apply on both strategic and operational management levels that are shaping supply chain network designs.