This thesis explores and evaluates the forecasting application of the machine learning method Gradient Boosting Decision Trees. This method is used to forecast the demand of the online grocery market with a 7-day time horizon. The thesis was conducted in collaboration with the online grocery company Mathem. The model is applied and evaluated on three different periods representing the spring, summer and fall. The main evaluation metric is the mean absolute percentage error (MAPE), and clear differences were found depending on the predictability of the period. Apart from the model and its application to demand forecasting, the related risk was investigated. This was done by studying the Value-at-Risk and Expected Shortfall associated with discrepancies between the forecasted and actual values over the three periods. The most important conclusion of the case study at Mathem is that overestimation in the forecast is more costly in terms of monetary value than underestimating. It is also found that this is highly dependent on the cost structure of the company's operation and could therefore vary between companies. Thus, the study has contributed to understanding the applications of machine learning models in forecasting processes as well as the risks related to over/underestimating the demand of the online grocery market. / I denna uppsats utforskas och utvärderas maskininlärningsmetoden Gradient Boosting Decision Trees och dess tillämpningsområde inom prognostisering. Metoden används för att prognostisera efterfrågan på onlinehandel av dagligvaror med en 7 dagars tidshorisont. Uppsatsen gjordes i samarbete med företaget Mathem som är aktiva inom denna sektor. Modellen appliceras och utvärderas på tre olika tidsperioder som representerar våren, sommaren och hösten. Modellen själv utvärderas med avseende på måttet mean absolute percentage error (MAPE) och tydliga skillnader mellan tidsperioderna observerades relaterat till variansen i datan. Förutom modellen och dess applikation på prognostisering av efterfrågan utforskades även de relaterade riskerna. Detta genomfördes genom att studera riskmåtten Value-at-Risk och Expected Shortfall baserade på differensen mellan de prognostiserade och de faktiska värdena under dessa tre tidsperioder. Den viktiga slutsatsen som kunde dras från uppsatsen i samband med Mathem var att det, i monetära termer, är dyrare att prognostisera för högt än för lågt. En ytterligare slutsats som kan dras är att den monetära risken är mycket beroende av kostnadsstrukturen hos företaget och skulle därför kunna variera mellan olika företag. Således har denna uppsats bidragit till förståelse för användandet av maskininlärning för prognostisering samt riskerna kopplade till över- och underskattning av efterfrågan i marknaden för dagligvaruhandel online.

Detaljhandeln befinner sig i ett skede av förändring som en effekt av de stora digitala framstegen som skett de senaste åren. Rapporter från fastighetsvärlden vittnar om mängder av butiker och kedjor som tvingats till nedskärningar och i extrema fall, konkurs. Förändring innebär både möjligheter och risker och den stora frågan är hur en ska agera för att lyckas minimera riskerna och maximera möjligheterna. Ett segment ur detaljhandeln har inte drabbats i samma omfattning, dagligvaruhandeln vilken innefattar livsmedelsbutiker som generellt fortfarande positiva siffror i den fysiska handeln. Det primära syftet med studien är att hitta den mest lämpliga hyresutformningen i butiksfastigheter, givet de mest förekommande hyresformerna utifrån hyresgästens perspektiv. Ett särskilt fokus har lagt på omsättningstillägg och dess lämplighet men indexerad bashyra undersöks också i fallstudien som jämförelseobjekt. Omsättningshyran och omsättningstillägget sägs ha haft sitt genomslag i Sverige under 1990-talets fastighetskris som en räddning av dåtidens höga vakansgrader. Metoden som används har varit av kvalitativ karaktär, där semistrukturerade intervjuer har genomförts med respondenter som representanter hyresgästens perspektiv med undantag för en respondent. Det korta svaret på vad som ingår i det optimala hyresavtalet bland ICA-handlarna är enstämmigt, en låg hyra. Undersöks saken mer ingående finns andra aspekter som är av stort värde så som en god dialog med fastighetsägare, låga vakansnivåer i centrumet och att fastighetsägaren ser långsiktigt på sitt ägande. ICA-handlarna som har en hyra med omsättningstillägg saknar ett tak på omsättningstillägget för att nå ett optimalt hyresavtal. Studien visar också att utformningen på ett optimalt hyresavtal är subjektiv och ser olika ut från fall till fall. Vägledande i valet av hyresform kan vara att omsättningstillägg endast är aktuellt i fall där en hyresgäst kan dra nytta av omkringliggande butikers kundflöden och centrumledning jobbar aktivt med centrumets utveckling. Finns inte detta så bör hyresgästen yrka på en fast hyressättning. Grundläggande för ett fungerade hyresavtal är att föra en dialog regelbundet med fastighetsägaren och ha en omfattande gränsdragningslista. / Retail is in the middle of a structure change as an effect of the major digital development taking place. The real estate industry reporting of numerous retailers that has been forced into cuts and in extreme cases, bankruptcy. Change means both opportunities and risks where the million dollar question is how to succeed in minimizing the risks and maximizing the opportunities. A segment that hasn´t been affected in the same extent is FMCG (=Fast moving consumer goods) that is still presenting positive figures from sales in the physical store. The purpose of this study is to find the most optimal rental arrangement in retail properties according to tenants, given the most prevalent rental forms. A special focus has been allocated to turnover rents and its suitability, indexed base rent is also analyzed for comparison purpose in the case study. The turnover rent is said to have entered the Swedish real estate market during the crisis of the 1990s as a rescue of the high vacancy rates occurring at the time. The method used has been qualitative, in which semi-structured interviews have been conducted with respondents who represent the tenant's perspective, one respondent represented the property owners perspective. The short answer to what is an optimal rental arrangement is among ICA retailers is unanimous, a low rent. If the matter is analyzed further in detail, there are other aspects of great value to the tenants such as a good dialogue with property owners, low vacancy rates and that the property owner are long-term owners. The ICA retailers who have a lease with a turnover rent are lacking a cap on the turnover rent in order to achieve an optimal rental arrangement. The study also shows that the optimal rent structure is subjective and vary from case to case. Guidelines in the rent negotiation may be that turnover rent is are only preferred in malls and cases where a tenant can benefit from the surrounding stores' footfall and also, when the shopping center management works actively with developing the mall. If these prerequisite are not fulfilled, the tenant should demand a fixed rent. Fundamental in all arrangements is to keep a dialogue regularly with the property owner and have an extensive list including delimitation of responsibility in the lease agreement.

The purpose of this thesis is to research how consumer recommendations can be used in terms of marketing and sales purposes in the Fast Moving Consumer Goods sector. Due to technological changes and a shift from traditional to digital marketing, the way of communicating to the audience has changed. Consumer recommendations, micro influencers and social media platforms are a part of digital marketing that is growing progressively. The paradigm of the thesis is interpretivism with following a qualitative research method, inductive reasoning and a case study in focus. The findings show that digital marketing, thanks to aspects as social media and e-commerce, has been growing, however has not completely taken over the marketing field. Influencers are getting a bigger role as marketing tools with the shift of power evolving. Due to these reasons is why and how consumer recommendations have become such a big trend nowadays.

Today, we can observe a shift in consumer behavior where more products are being consumed online. For food or so called fast moving consumer goods it has taken longer to start this transition than for other industries such as books and electronics but now it is on the rise and it has been an overall increase of 30 to 37 % in the last couple of years. A continuous increase in e-commerce will most likely lead to major changes in the real estate industry as the retail industry will not be in need of physical stores to the same way as they are used today. In line with these structural changes new business models will be created that strikes out old ones.  The aim of this report is to account for the development of e-commerce within fast moving consumer goods (FMCG) and to evaluate the business models of the main traditional retailers. By comparing how retailer’s business models meet the underlying theory of consumers purchasing behavior online, the purpose is to explain what retailers will be successful in the future. To fulfil the aim of the report an examination has been made of relevant theories such as the theory of planned behavior and its key factors essential to run a successful business model for online retailing within FMCG. Apart from the theory of planned behavior, the business model of IKEA that has been proven to be successful, is used as a reference model. The research is restricted to two of our four main retailers in Sweden for food which is ICA and Bergendahls (City Gross). To answer the research questions a qualitative method has been used in the thesis by performing two case studies on ICA respectively Bergendahls with the collection of primary data in form of interviews, which have been complemented with secondary data. By comparing the business models of ICA and Bergendahls with the theoretical models and the reference object IKEA the purpose is to determine if they will be successful online or not. The key to having a successful business model and online platform is to focus mainly on the wired lifestyle and the customers attitude towards it, which is the most important factor affecting the consumers’ intention towards buying food online. Freshness and novelty are the second and third most important factors affecting the consumers’ intention towards buying food online and therefore shall these factors be the retailers next focus area. In order to pin point what is important for their business and to be able to focus on the right factors for being successful and gaining market shares the retailers should have a separate model for e-commerce. / Idag kan vi observera ett förändrat konsumentbeteende där fler produkter konsumeras online. För mat som hör till så kallade snabbrörliga konsumtionsvaror har det tagit längre tid att påbörja denna övergång mot e-handel i jämförelse med andra industrier så som böcker och elektronik, men denna förändring är nu på god väg och det har skett en övergripande ökning med 30 - 37% under de senaste åren. En kontinuerlig ökning av e-handel kommer sannolikt att leda till stora förändringar inom fastighetsbranschen eftersom att detaljhandeln inte kommer att behöva ha fysiska butiker i samma utsträckning. I linje med dessa strukturella förändringar skapas nya affärsmodeller som slår ut gamla. Syftet med denna rapport är att redogöra för utvecklingen av e-handel inom snabbrörliga konsumentvaror (FMCG) och att utvärdera affärsmodellerna hos de största detaljhandlarna. Genom att jämföra hur detaljisternas affärsmodeller uppfyller den underliggande teorin om konsumenters köpbeteende online, är syftet att förklara vilka återförsäljare som kommer att lyckas i framtiden sett till den nämnda teorin. För att uppfylla syftet med rapporten har relevanta teorier granskats så som teorin om planerat beteende och dess centrala faktorer som är nödvändiga för att driva en framgångsrik affärsmodell inom online-detaljhandel för snabbrörliga konsumentvaror. Förutom teorin om planerat beteende används affärsmodellen för referensobjektet IKEA som referensmodell på grund av IKEA:s framgång.  Forskningen är begränsad till två av de fyra största återförsäljarna av mat i Sverige vilka är ICA och Bergendahls (City Gross). För att besvara forskningsfrågorna har en kvalitativ metod använts i avhandlingen genom att utföra två fallstudier på ICA respektive Bergendahls med insamling av primära data i form av intervjuer, vilka kompletterats med sekundära data. Genom att jämföra ICA:s och Bergendahls affärsmodeller med de teoretiska modellerna samt med referensobjektet IKEA är syftet att bestämma om återförsäljarnas affärsmodeller kommer att bli framgångsrika online eller ej. Slutsatsen som nåtts är att nyckeln till att ha en framgångsrik affärsmodell och online-plattform är att fokusera främst på den uppkopplade livsstilen och kundernas inställning till den, vilket är den viktigaste faktorn som påverkar konsumenternas intention till att köpa mat online. Färskhet och nymodighet är den andra och tredje viktigaste faktorn som påverkar konsumenternas intention till att köpa mat online och därför bör dessa faktorer vara detaljhandlarens nästa fokusområde. Detaljhandlarna bör ha en separat affärsmodell för e-handel för att kunna precisera det viktigaste för deras affärer, kunna fokusera på rätt faktorer för att nå framgång och ta marknadsandelar.

This thesis investigates employee attitudes to corporate citizenship (CC) and the impact on employer brands. It addresses the practice of CC in the German FMCG industry, considers employee understanding of CC, and examines which CC initiatives influence perceptions of employer brands. The research is driven by the dual challenges of sustainable development and the ‘war for talent’ in attracting, motivating and retaining employees. It is underpinned by the extant literature on CC dimensions, stakeholder theory and employer brands. The research approach is based on two case studies, comprising three phases: analysis of corporate documentation followed by two phases of semi-structured interviews exploring employees’ perceptions of CC and the link to employer brands. Key findings: An understanding of the positioning of CC in the corporate sustainability strategy and differing foci of CC, including environment, sustainable supply chain and people/culture. The study highlights employee understanding of eight CC dimensions, revealing economic responsibility toward employees as a new dimension. With respect to employer brands, CC initiatives focused on discretionary responsibility towards the natural environment and economic responsibility towards customers are seen as essential; legal, ethical and discretionary responsibility towards community are limited; and discretionary and economic responsibilities towards employees have a strong relationship. A focus on economic responsibilities towards owners has a strong but negative relationship. The theoretical contribution is a conceptual framework of all identified CC dimensions in practice and their relationship to employer brands. Contributions to practice include the importance of benefit packages, work-life balance support, employee development and work environment.

FMCG快速移動消費性產品泛指商品使用週期較短、產品消費速度較快、消費者需要不斷重複購買的產品，由於消費者對此種產品的涉入程度低、認知品牌間產品特性差異不大，為了提高市佔率，企業常會以品牌價值創造做為行銷要點，因此，本研究主要依據邱志聖教授所提出的「4C策略行銷分析架構」，就FMCG買方面對品牌廠商所投入的行銷組合的價值評估，與賣方在交易過程獲得品牌權益累積的作法，以雙方所投入的四種成本：外顯單位效益成本 (C1)、資訊搜尋成本(C2)、道德危機成本(C3) 以及專屬陷入成本(C4) ，找出品牌發展與建構的成功關鍵因素。 採用居家領導品牌案例並針對產品生命週期的產品導入期、成長期等品牌建構、發展時期進行成功因素分析，發現： 1.就產品導入、品牌建構期： 品牌建構的基礎是商品，品牌廠商必須確保消費者獲得與品牌承諾一致的C1商品品質與效益，產品的開發策略也必須聚焦於產品本身的關鍵技術或者利基開發滿足各種消費者需求，且能不斷提升與累積其使用產品總效益的產品組合延伸，再透過有效的C2投資，建議採用中央路徑直接溝通品牌的商品獨特利益，並採用整合性的行銷傳播媒體傳遞一致的商品訊息與品牌形象，建立一致清楚的品牌內涵並系統性的累積品牌資產。品牌建構的過程必須在商品上市前即對於可能的C3成本建立有效的管控做法與危機處理機制，而持續透過與買者C1效益連結為品牌商品的專屬資產，透過C2活動建立的品牌印象累積品牌專屬資產C4，建立品牌忠誠。 2.就產品成長、品牌發展期： 維持品牌商品的穩定產品品質與運用關鍵技術進入新區隔市場或者產品線延伸都是持續降低消費者外顯單位效益成本C1的有效做法，再透過活用不同的溝通通路與新興科技整合傳遞一致的定位與商品、品牌訊息，降低消費者C2並持續累積品牌C1效益訊息傳遞進而建立品牌的專屬資產C4。品牌發展階段C3的管理控制作業甚或危機處理致關重要，所提及穩定的C1產品品質、透過C2建立C4品牌信任感、面對品牌競爭的C2回應操作都是控制或降低道德危機成本C3的關鍵因素。自品牌建構期到發展期必須定義清楚與系統性的累積消費者對於該品牌的技術方案產品效益、品牌印象偏好等品牌的專屬陷入成本C4，持續的追蹤與運用C4創造更深厚的品牌資產，產生態度與行為的忠誠，促使品牌永續的發展。

The purpose of the presented project is to eliminate the supply chain losses in Kraft Heinz Company through preventive actions. The two points that the current thesis will focus on are the prevention of excess production and the elimination of MOQ (Minimum Order Quantity) issues that contribute to supply chain losses. High MOQ is one of the main root causes of excess production. Firstly, a literature review will be made. The approach of creating the literature framework is to start from a general description of supply chain activities in FMCG (Fast-moving Consumer Goods) companies, then to focus on more specific topic which is the excess stock and the inventory management and finally, to end up in the food waste hierarchy analysis that scientific articles frame. Food waste for a FMCG company is similar to supply chain losses, thus it is critical to provide a state-of-art literature in order to have a better understanding. Furthermore, the KHC (Kraft Heinz Company) background will be described, as the history of KHC and the current portfolio in order to understand the size of the company and the needs of such a huge company and the complexities. Additionally, the supply chain function in KHC will be described in details in order to understand the reason of this thesis existence, to identify the gaps in this process and the improvements should be made. Finally, the improvements made through this project will be presented, as well as, the processes and the methods that are developed and standardized. The analysis of the prevention of excess production process, as well as, the improvements have been made in this process will be explained. The routines and the new approaches will be described. The second part of the improvements section is concerned the development and the standardization of the MOQ Flexibility project. The calculations, the process, the challenges and the involved stakeholders will be part of this thesis. / Syftet med det presenterade projektet är att eliminera förlusterna i försörjningskedjorna genom förebyggande åtgärder på företaget Kraft Heinz. De två delar som examensarbetet fokuserar på är hur man ska kunna förhindra överproduktion och eliminera MOQ (Minimum Order Quantity/(Minsta orderkvantitet) faktorer som bidrar till förlust i försörjningskedjan. Hög MOQ är en av de främsta orsakerna till överproduktion. Det första steget var att göra en litteraturundersökning. Litteraturöversikten skapar ett ramverk där utgångspunkten är en allmän beskrivning av aktiviteter i leverantörernas försörjningskedjor inom företaget för FMCG (Fast Moving Consumer Goods/ Snabbt Rörliga Konsumentvaror), för att sedan fokusera på ett mer specifikt ämne, vilket är överskottslagret och lagerstyrningen för att slutligen landa i en matavfallshierarkianalys som ramas in av vetenskapliga artiklar. Matavfall för ett FMCG- företag är samma sak som försörjnings- kedjeförluster i leverantörskedjan, därför är det viktigt att tillhandahålla en nulägesanalys med referens till litteraturen för att få en bättre förståelse. Dessutom beskrivs KHC (Kraft Heinz Company) mot bakgrund av företagets historia och den nuvarande produktportföljen för att läsaren ska få en uppfattning om företagets storlek, dess behov och komplexiteten i ett sådant stort företag. Leveranskedjefunktionen inom KHC kommer beskrivas i detalj för att man ska kunna förstå anledningen till behovet av detta examensarbete, för att identifiera luckorna i denna process och för att peka på de förbättringar som bör göras. Slutligen kommer de förbättringar som gjorts genom detta projekt att presenteras, liksom, processerna och metoderna som utvecklats och standardiserats. Analysen med syfte att förhindra överskott i produktionsprocessen samt de förbättringar som har implementerats i denna process kommer att förklaras. Rutinerna och de nya tillvägagångssätten kommer att beskrivas. Den andra delen av förbättringsavsnittet handlar om utveckling och standardisering av MOQ Flexibilitetsprojektet. De beräkningar, processen, utmaningarna och de inblandade intressenterna är en del av denna avhandling.

Thesis (MBA)--Stellenbosch University, 2007. / ENGLISH ABSTRACT: Product Mix Complexity is a very real issue in modern companies. As globalisation and competition increase, markets mature and the needs of individual consumers get ever more specific, the trend towards more product variety will grow. Product Mix refers to the full range of products that a company offers to its customers. Product Mix Complexity refers to the effect that this product variety has on organizational performance. Understanding and management of this problem is difficult due to the many organizational elements involved, the complicated relationship between these elements and the fact that the problem crosses functional organization boundaries. A company may choose to differentiate itself in the market by offering a broad product mix. To do this effectively the price that can be asked for an item must offset the additional costs brought on by this complexity. This balance between variety that drives sales, and the costs of the added complexity is at the core of the effective management of Product Mix Complexity. The effect of this complexity in manufacturing relates to the loss of scale efficiencies and the need for flexibility in operations. Due to the difficulty in understanding the complex effects of variety, company portfolios tend to proliferate, resulting in a very skewed spread of product contributions where a small % of products contribute a high proportion of company profits. Management processes that continuously evaluate the total profitability of their portfolios from a ‘cost of complexity’ point of view are needed. Cost accounting systems often do not accurately account for this cost of complexity. The effects of Product Mix Complexity are investigated in Cadbury South Africa. The business displays clear signs of having a classical ‘Pareto’ spread of products where a ‘long tail’ of small volume products add very little profit to the business. The Port Elizabeth factory is an above average complex plant within the Cadbury group. It was found that the cost allocation system employed by Cadbury is not accurately allocating costs to products and is thereby aggravating the proliferation of the Cadbury product portfolio. Potential savings by rationalizing the Cadbury product portfolio are identified. A number of recommendations to better manage the presence of Product Mix Complexity are made, both for the business as a whole and for manufacturing specifically. / AFRIKAANSE OPSOMMING: Produkmengselkompleksiteit is alomteenwoordig in moderne besighede. Die neiging tot meer produkverskeidenheid groei namate globalisering, kompetisie en verbruikers se behoeftes na verskeidenheid toeneem. ‘Produkmengsel’ verwys na die volle reeks produkte wat ’n maatskappy aan die mark bied. ‘Produkmengselkompleksiteit’ verwys na die effek wat produkverskeidenheid het op die doeltreffendheid van ’n besigheid. Die probleem is moeilik om te definieer en te verstaan as gevolg van die baie besigheidselemente wat betrokke is, die komplekse verhoudinge tussen hierdie elemente en die multi-funksionele aard van die probleem. Vir strategiese redes wil ’n maatskappy soms ’n groot verskeidenheid produkte aanbied om hom te onderskei in die mark. Om hierdie strategie winsgewend te maak moet die prys wat behaal word opmaak vir die ekstra kostes wat aangegaan word om die breë verskeidenheid te kan aanbied. Die balans tussen die interne koste van verskeidenheid en die voordeel van die verskeidenheid is baie belangrik in die effektiewe bestuur van produkmengselkompleksiteit. Omdat die effek van verskeidenheid moeilik is om te definieer, neig maatskappye daarna om liewer te veel as te min produkte aan te bied. Die gevolg is dikwels ‘n ongebalanseerde distribusie van produkwinsgewendheid waar ’n klein persentasie produkte verantwoordelik is vir die oorgrote meerderheid van die wins. Bestuursprosesse wat produkportefeuljes vanuit ‘n kompleksiteitskoste oogpunt evalueer ontbreek dikwels. Produkkostemodelle neem ook dikwels nie hierdie koste akkuraat in ag nie. Die effek van produkmengselkompleksiteit in Cadbury Suid-Afrika word in hierdie studie ondersoek. Dit is duidelik dat Cadbury Suid-Afrika ’n klassieke ‘Pareto’ effek vertoon waar ’n lang stert van klein produkte baie min tot besigheidswins-gewendheid bydra. Die Port Elizabeth fabriek vertoon ook bogemiddelde kompleksiteit relatief tot ander Cadbury fabrieke. Kompleksiteitskostes word nie akkuraat toegedeel tot produkkostes nie, met die gevolg dat klein produkte aanhoudend tot die produkmengsel gevoeg word. Die potensiële besparings as gevolg van rasionalisasie van die produkportefeulje is bereken. Aanbevelings vir die beter bestuur van produkmengselkompleksiteit word gemaak vir die besigheid as ’n geheel en vir vervaardiging spesifiek.

Consumer marketing scholars keenly emphasize a proposed paradigm shift toward interactive relationships and lived brand experiences. Yet, little has been done to investigate the link between the two. Until now. This study is an attempt to measure the effects of lived brand experiences on consumers’ perceived relationship with a brand, through testing an academically established brand relationship quality model onto the concept of experiential event marketing. Susan Fournier’s (2000) brand relationship quality scale was chosen as the construct to be tested in the experiential event marketing context. It was through a theoretical argumentation hypothesized that the experiential event intervention would produce positive direct effects within the scale, but that these would decline with time. This was consequently tested through a repeated measurement study, set at an experiential food truck event hosted by the Swedish FMCG brand Santa Maria. Respondents were to rank their perceived brand relationship quality with the brand on three different occasions; directly before, directly after, and two weeks after being exposed to the experiential event. This way, not only the immediate effect, but also the effect over time, could be measured. It could be concluded that all but one constructs produced positive direct effects, but only half of them were significant. In all cases but one this effect declined significantly when being measured two weeks afterwards, and went in several cases back at approximately the same level as in the initial measurement. These findings have important implications for both academics and practitioners. Most notably, we argue that the link between lived brand experiences in form of typical FMCG experiential events and strengthened longer-term brand relationship quality can be invalidated.

Na área de processamento de sinais biomédicos a extração de informação, baseada em um conjunto de medidas adquiridas no tempo, é considerada de suma importância. A qualidade desta informação extraída permite avaliar o funcionamento dos diversos órgãos. Objetivos: (1) propor o método de análise de componentes dependentes para auxiliar a extração de componentes de interesse, a partir de medidas multivariadas; (2) caraterizar as componentes extraídas através de representações em termos de tempo e freqüência, e espectro de potência; e, (3) aplicar o método e avaliar as componentes de interesse extraídas no contexto real MCGf, MGG e fMRI. A proposta para a extração fundamenta-se no método chamado de Análise de Componentes Dependentes ACD. As medidas a serem processadas são multivariadas a partir de sensores distribuídos, espacialmente, no corpo humano dando origem a um conjunto de dados correlacionados no tempo e/ou no espaço. Observa-se que os sinais de interesse raramente são registrados de forma isolada, e sim misturados com outros sinais superpostos, ruído e artefatos fisiológicos ou ambientais, onde a relação sinal-ruído é geralmente baixa. Nesse contexto, a estratégia a ser utilizada baseia-se na ACD, que permitirá extrair um pequeno número de fontes, de potencial interesse, com informações úteis. A estratégia ACD para extração de informação é aplicada em três importantes problemas, na área de processamento de sinais biomédicos: (1) detecção do sinal do feto em magnetocardiografia fetal (MCGf); (2) detecção da atividade de resposta elétrica do estômago em magnetogastrografia (MGG); e, (3) detecção de regiões ativas do cérebro em experimentos de imagens por ressonância magnética funcional (Functional Magnetic Resonance Imaging, fMRI). Os resultados, nos três casos estudados, mostraram que o método utilizado, como estratégia, é efetivo e computacionalmente eficiente para extração de sinais de interesse. Concluímos, baseados nas aplicações, que o método proposto é eficaz, mostrando seu potencial para futuras pesquisas clínicas. / An important goal in biomedical signal processing is the extraction of information based on a set of physiological measurements made along time. Generally, biomedical signals are electromagnetic measurements. Those measurements (usually made with multichannel equipment) are registered using spatially distributed sensors around some areas of the human body, originating a set of time and/or space correlated data. The signals of interest are rarely registered alone, being usually observed as a mixture of other spurious, noisy signals (sometimes superimposed) and environmental or physiological artifacts. More over, the signal-to-noise ratio is generally low. In many applications, a big number of sensors are available, but just a few sources are of interest and the remainder can be considered noise. For such kind of applications, it is necessary to develop trustful, robust and effective learning algorithms that allow the extraction of only a few sources potentially of interest and that hold useful information. The strategy used here for extraction of sources is applied in three important problems in biomedical signal processing: (1) detection of the fetal magnetocardiogram signal (fMCG); (2) detection of the electrical activity of the stomach in magnetogastrograms (MGG); and (3) detection of active regions of the brain in experiments in functional Magnetic Resonance Imaging (fMRI). The results, within the three cases of study, showed that the DCA method used as strategy is effective and computationally efficient on extraction of desired signals.