Målet med detta projekt var att optimera produktionslinjen för att maximera antalet färdigställda artiklar som produceras. Projektet påbörjades med en planering och en undersökning av de verktygen som skulle användas så som simulering, lean och information relaterat till produktionslinjen. Därefter modellerades den i simulerings-verktyget utifrån den faktainsamlingen som utfördes och simuleringsmetoden. När modellen var klar utfördes experimentella tester för att undersökningar hur specifika faktorer så som strategiskt placerade buffertar och antalet operatörer påverkade effektiviteten på produktionslinjens bearbetning. Utifrån resultaten av dessa tester bestäms en optimal nivå för faktorerna som skall användas i full factorial design testerna. Med dessa tester utförda kunde det noteras att ingen optimering kunde utföras på produktionslinjen då simuleringen uppnådde endast 72% när målet för produktionslinjen är 75%. Istället gavs andra strategiska rekommendationer på vad företaget kunde göra för att öka effektiviteten på produktionslinjen. Dessa rekommendationer inkluderar, en tillfällig buffert, vänta ut de tillfälliga stoppen, åtgärda avvikande processteg och minska kassationerna. / The goal of this project was to optimize the production line to maximize the number of completed items produced. The project started with a planning and research of the tools which will be used such as simulation, lean principles, and information related to the production line. Thereafter, it was modelled in the simulation tool based on the data collection that was carried out and the simulation method. Once the model was ready, experimental tests were conducted to investigate how specific factors, such as strategically placed buffers and the number of operators, affected the efficiency of the production line's processing. Based on the results of these tests, an optimal level for the factors to be used in the full factorial design tests was determined. With these tests performed, it was noted that no optimization could be carried out on the production line as the simulation achieved only 72% efficiency when the target for the production line is 75%. Instead, other strategic recommendations were provided on what the company could do to increase the efficiency of the production line. These recommendations include, a temporary buffer, waiting out temporary stoppages, addressing deviant process steps, and reducing rejections.

Ti-6Al-4V has a wide range of applications such as in the automotive and aerospace industries. Nevertheless, titanium alloys are very difficult to machine by conventional methods. Micro-EDM is a non-conventional machining method that uses the thermal effect of precisely controlled sparks. Manufacturers are looking for the methods and optimal settings to increase the productivity of micro-EDM in terms of lessening machining time and tool wear. Moreover, surface integrity of the machined area is crucial for some products such as biomedical implants. The objective of this study was to investigate the effects of the micro-EDM process parameters on response variables, in order to understand the behavior of each parameter as well as to determine their optimal values. Although, there is a substantial amount of literature studying different aspects of micro-EDM, most of them were designed based on the one-factor-at-a-time experiments instead of studying all factors, simultaneously. This research was conducted through a series of experiments using a full factorial design. An analysis of variance was employed to analyze the findings and to determine the effect and significance of each process parameters on the response variables. The process parameters included voltage, capacitance, electrode rotational speed, and electrode coating. Voltage and capacitance were studied separately as well as in combination in terms of the discharge energy. Response variables consisted of machining time, tool wear, crater size, microhardness, and element characterization. The surface morphology and element characterization were studied through the application of SEM and EDS analysis. The findings indicated that voltage had a decreasing effect on machining time, while it increased the crater size. Capacitance decreased machining time and tool wear. It had an increasing effect on the surface hardness. The effects of the TN-coating and electrode rotational speed were not statistically significant. Voltage and capacitance were the only parameters affecting element characterization. The affected elements included Ti, Al, C, and W. The optimal process parameters for two sets of response variables were determined using Minitab 17.

Through a full factorial design experiment, the effects of time between worts, wort aeration and yeast dosage in production of a pale lager beer were examined in the beer process at Spendrups Bryggeri AB. The aim was to learn how different parameters may affect the yeast fermentation performance during beer production. Response variables used were the concentrations of ethyl acetate and isoamyl acetate, free amino nitrogen (FAN) degradation and change in extract. A statistical analysis showed that the concentration of ethyl acetate is dependent on yeast dosage and the interaction between time between worts and aeration while the isoamyl acetate concentration is dependent on yeast dosage and time between worts. No parameters are statistically significant for FAN degradation while the change in extract is dependent on the yeast dosage. Due to botched runs, mostly because of aeration problems, it was not possible to verify theoretical parameter values and responses. Since the aeration was not properly performed, the management of the aeration control should be further investigated. Ester analysis and analysis of FAN were performed as worts entered and exited horizontal fermentation tanks. An additional analysis of ester content was also performed as the early stage beer was transferred into lagering tanks. Cell viability as well as extract, pH and tank temperature was measured daily to verify the state of fermentation. Statistical calculations showed that when using NucleoCounter YC-100, there is no significant difference between analysis made of samples homogenized by a magnetic stirrer and samples shaken by hand.

Hydrothermally treated biomass could not only be used as a fuel or a fertilizer but it can also be refined into high-value products. Activated carbons are one of those. In the studies of this thesis, four different hydrothermally carbonized (HTC) biomasses, including horse manure, grass cuttings, beer waste and biosludge, have been successfully made into activated carbons. The activated carbon materials were in the forms of powdered activated carbons, powdered composites of activated carbon and iron oxide nano-crystals, and activated carbon discs. The HTC biomasses and the activated carbons were characterized and analyzed using several methods. The biomasses were carbonized to different extent during the hydrothermal treatment, which depended on the different natures of the biomasses. The HTC biomasses were activated into powdered activated carbons by both physical activation, using CO2, and by chemical activation, using H3PO4. Full factorial design matrices were applied to design experiments and study the influence of different parameters used during both physical and chemical activation. Activated carbons with embedded iron oxide nanoparticles were synthesized through hydrothermal carbonization followed by CO2 activation. These composites had high surface areas and showed a strong magnetism, and the powders could be separated from liquid phase by applying a magnetic field. Strong and dense activated carbon discs were also prepared from powdered HTC beer waste by pulsed current processing (PCP) and a subsequent CO2 activation procedure. The potential for carbon dioxide separation from nitrogen, and methylene blue adsorption in aqueous solution, were assessed for the powdered activated carbons produced from HTC biomasses. They showed good performance in both applications. / <p>At the time of the doctoral defense, the following paper was unpublished and had a status as follows: Paper 1: In press.</p>

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O objetivo deste trabalho é a aplicação de um Planejamento Fatorial Completo, ferramenta estatística que auxiliará a obter dados empíricos para otimizar um tratamento termoquímico, a cementação. Partindo-se de um levantamento da profundidade de camada cementada e da dureza de uma corrente de aço de baixo teor de carbono, usada para amarração, e para resistir à abrasão reproduziu-se uma nova cementação, variando-se seus parâmetros de influência, para se alcançar o ponto ótimo, definindo o melhor aço e o melhor processo. Foram realizados dois planejamentos, um fatorial 2 e dois 2, comparando o comportamento do processo na prática em relação aos resultados teóricos de uma simulação computacional, que permite a obtenção das curvas de enriquecimento de carbono, baseado na segunda Lei de Fick, para várias condições de contorno. Os perfis teóricos de cementação apresentaram valores de profundidade efetiva próximos aos valores obtidos experimentalmente, evidenciando o planejamento realizado.

Made available in DSpace on 2016-12-08T17:19:36Z (GMT). No. of bitstreams: 1 PRE-TEXTUAIS.pdf: 48578 bytes, checksum: 21f78cab7ee8f139f12bf814c8bfa87f (MD5) Previous issue date: 2010-07-29 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / This work describes the results of an investigation on the possible applications of ground waste calcined clay bricks (WCB) for use as an active partial replacement for Portland cement in mortars. The investigation has been carried out using a full factorial design for experimental design. For the study, the physical and mechanical properties along their durability of mortar produced with WCB were investigated and the results presented. Mortar mixtures were prepared using WCB as the partial replacement of Portland cement at the level 10%, 20% and 30 wt.% with water/cement ratios of 0.45, 0.55 and 0.65. The necessary data for establishing a mix proportion design and a quality control method are obtained. Properties of fresh (consistency index and specific gravity) and hardened (7-day, 28-day and 120-day compressive strength, 28-day water absorption and 28-day alkali content) mortars were evaluated on specimens. Such statistical modeling is discussed against qualitative X-ray diffraction and microstructures of some selected samples were studied using SEM. Several empirical equations, surface and contour plots for the physicalmechanical properties of fresh and hardened recycled mortar with WCB as replacement content were obtained via regression analysis. The influence of level of WCB on the properties has been established, and its suitability for use in a mortar application has been assessed. The consistency decreased in parallel to an increase in the proportion of WCB. Specific gravity of mortars was far lower than that of normal no replacement mortars. A given 28-day compressive strength (21 MPa to 37 MPa) can be achieved at several water/cement ratio and the replacement contents up to 30 wt. %, thus producing na economical mixture mortar with WCB and lower cement Portland content. XRD and SEM analyses results indicated the presence of quartz, calcite, ettringite, portlandite, and silicates in recycled mortars. / Este trabalho apresenta os resultados da investigação do uso de rejeitos particulados de tijolos (RPT) na forma de pó, proveniente de tijolos de cerâmica vermelha, como substituinte parcial do cimento Portland, na produção de argamassas para construção civil. O estudo foi realizado adotando-se o projeto fatorial completo 32 acoplado com técnicas de metodologia de superfícies de resposta. Foram preparadas argamassas usando RPT em níveis de 10%, 20% e 30% (em massa), em substituição ao cimento Portland, com os níveis de relação água/cimento de 0,45, de 0,55 e 0,65. As misturas foram curadas por períodos de 7, 28 e 120 dias. As propriedades das argamassas contendo RPT no estado fresco foram avaliadas: massa específica e índice de consistência. As propriedades avaliadas das argamassas com RPT no estado endurecido foram: resistência à compressão (7, 28 e 120 dias), absorção de água aos 28 dias e teor de álcalis (CaCO3) aos 28 dias. Os resultados obtidos através da metodologia estatística foram confrontados com as análises de difração de raios X (DRX) e microscopia eletrônica de varredura (MEV) de amostras selecionadas. Foram obtidas equações empíricas, gráficos de superfície de resposta e de contorno para as propriedades físico-mecânicas das argamassas com RPT no estado fresco e endurecido, por meio de análise de regressão. A consistência diminuiu linearmente com o aumento da quantidade de RPT incorporado em substituição ao cimento Portland. A massa específica das argamassas contendo RPT foi menor do que na argamassa sem RPT. Foi possível obter argamassas com resistências à compressão aos 28 dias de 21 MPa a 37 MPa para vários níveis de relação água/cimento em até 30 % de substituição de cimento Portland por RPT. Dessa forma, é possível produzir argamassas econômica e ambientalmente viáveis, a partir da substituição parcial de cimento Portland por RPT, nas dosagens estudadas. As análises de DRX e de MEV indicaram a presença do quartzo, da calcita, da etringita, da portlandita, silicatos e porosidade em amostras de argamassas selecionadas, de modo que foi possível relacionar as mesmas com a resistência à compressão das argamassas.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O objetivo deste trabalho é a aplicação de um Planejamento Fatorial Completo, ferramenta estatística que auxiliará a obter dados empíricos para otimizar um tratamento termoquímico, a cementação. Partindo-se de um levantamento da profundidade de camada cementada e da dureza de uma corrente de aço de baixo teor de carbono, usada para amarração, e para resistir à abrasão reproduziu-se uma nova cementação, variando-se seus parâmetros de influência, para se alcançar o ponto ótimo, definindo o melhor aço e o melhor processo. Foram realizados dois planejamentos, um fatorial 2 e dois 2, comparando o comportamento do processo na prática em relação aos resultados teóricos de uma simulação computacional, que permite a obtenção das curvas de enriquecimento de carbono, baseado na segunda Lei de Fick, para várias condições de contorno. Os perfis teóricos de cementação apresentaram valores de profundidade efetiva próximos aos valores obtidos experimentalmente, evidenciando o planejamento realizado.

In the pharmaceutical industry, it is always important to have reproducible processes and raw materials of high quality to ensure good quality products. AstraZeneca, that is a leading manufacturer of different pharmaceuticals, works according to GMP to make sure that their processes deliver products of the same quality every time. A problem that has occurred at AstraZeneca is when a raw material is not properly understood and variations in the raw material affects the final product. Variations in drug release in one of AstraZeneca´s products, Product X, has been linked to the cushioning excipient Microcrystalline cellulose (MCC). Variations in drug release has been noticed during change from one batch of MCC to another. The aim of this study was to investigate which material attributes of MCC that contributes to variations in the final product. Particle size and moisture content were identified as critical material attributes (CMA´s) and were therefore chosen to be investigated more thoroughly. By variating particle size and moisture content during manufacturing of Product X, the influence of these attributes could be investigated using Design of Experiment (DoE). An additional experiment that compared two MCC batches from different suppliers was also performed during this study. The results from these experiments showed that the particle size and moisture content of MCC does affect the drug release. Large particles and high moisture content gave rise to a faster drug release compared to small particles and low moisture content that gave rise to a slower drug release. It is however hard to draw conclusions regarding how small differences in particle size and moisture content could affect the drug release.

Distributed control systems play a central role for critical processes within a plant that needs to be monitored or controlled. They ensure high production availability and output while simultaneously ensuring the safety of the personnel and the environment. However, 5% of global annual production is lost due to unscheduled downtime. 80% of the unscheduled shutdowns could have been prevented and 40% of these are caused by human error.  This study is conducted at ABB's Process Automation team in Umeå. The aim is to examine if different human-machine interfaces affect operators' effectiveness in resolving errors and maintaining a high production level. DoE is the chosen approach for this study which includes planning and conducting an experiment where the two dependent variables are Effect and Time. The independent variables examined are Scenario, Graphic, and Operator which are used as factors in a factorial design, each having two levels.  Experiments showed that the design of the human-machine interface has no impact on either responses, i.e. it has no statistically significant effect on the production in terms of operator effectiveness or production efficiency. Instead, the level of experience of the operators seems to be the main contributor of variance in production in the models used. / Distribuerade styrsystem spelar en central roll för kritiska processer inom en anläggning som måste övervakas eller kontrolleras. De säkerställer hög produktionstillgänglighet ochproduktion samtidigt som säkerheten för personalen och miljön säkerställs. Det har visats att 5% av den globala årsproduktionen går förlorad på grund av oplanerade driftstopp. 80% av de oplanerade avbrotten kunde ha förhindrats och 40% av dessa orsakas av den mänskliga faktorn. Denna studie genomförs hos ABB:s Process Automation team i Umeå. Målet är att undersöka om olika gränssnitt för styrsystemen är en viktig faktor för operatörens effektivitet i att åtgärda fel och att upprätthålla en hög produktionsnivå. Försöksplanering är det valda tillvägagångssättet för denna studie som inkluderar planering och genomförande av experimentet där de två beroende variabler är Effekt och Tid. De oberoende variabler som undersöks är Scenario, Grafik och Operatör, och används som faktorer i en faktoriell design, där faktorerna har två nivåer vardera. Experimentet visade att utformningen av den grafiska designen för gränssnittet inte har någon inverkan på någondera svaren, d.v.s. den har ingen statistiskt signifikant effekt på produktionen i form av operatörseffektivitet eller produktionseffektivitet. Istället tycks operatörernas erfarenhetsnivå vara den främsta orsaken till variationen i produktionen i de modeller som används.

Made available in DSpace on 2017-07-20T12:40:14Z (GMT). No. of bitstreams: 1 Giselle Nathaly Calaca.pdf: 4129912 bytes, checksum: 17ad0dde0faa7762fe24c34e6ffcd471 (MD5) Previous issue date: 2015-03-23 / Fundação Araucária de Apoio ao Desenvolvimento Científico e Tecnológico do Paraná / The development and validation of analytical methods for quantification of two associations, antibiotics: sulfamethoxazole (SMX) and trimethoprim (TMP); and depigmenting agents: kojic acid (AK) and hydroquinone (HQ), using unmodified glassy carbon electrode (ECV) and square wave voltammetry (SWV) are described. For the first association, two well defined irreversible oxidation peaks were obtained at 0.96V (SMX) and 1.12V (TMP) in Britton–Robinson buffer (pH 6.0), characterized by diffusion and adsorptive processes, respectively. SWV parameters were optimized by means of 23 Full Factorial Design, and the best analytical signal, in terms of sensitivity and selectivity is obtained at a=30 mV, f=100 s-1 and ΔES=5mV. Under optimized conditions, SWV measurements showed excellent linear response, from 5.5x10-5 to 3.95x10-4 mol L-1 (R=0.9971) and 1.05x10-5 to 1.04x10-4 mol L-1 (R=0.9974) for SMX and TMP, respectively. The detection and quantification limits were found to be 8.52x10-6 mol L-1 and 2.84x10-5 mol L-1 for SMX; and 9.31x10-7 mol L-1 and 3.10x10-6 mol L-1 for TMP, respectively. The proposed method was successfully applied to the simultaneous determination of these antibiotics in commercial pharmaceutical formulations (tablets, oral suspension and injection), without any sample pretreatment. The obtained results are in good agreement with that obtained by the standard HPLC method at a 95% confidence level. For the second association, the redox peak potentials were completely separated in Mcllvaine buffer solution (pH 8.0), KA exhibits a well-defined irreversible oxidation peak at 0.72V and HQ a quasi-reversible redox peak at 0.08 V and -0.02 V, characterized by adsorptive and diffusional processes, respectively. Under SWV parameters optimized (a= 40 mV, f=40 s-1 and ΔES=1mV) by 23 Full Factorial Design, the calibration curves were obtained in the concentration ranges of 3.5×10-5 to 2.5×10-4 mol L-1 (R= 0.9996) for kojic acid and 1.5×10-5 to 1.6×10-4 mol L-1 (R= 0.9994) for hydroquinone. The detection and quantification limits were found to be 7.84x10-6 mol L-1 and 2.61x10-5 mol L-1 for KA; and 3.71x10-6 mol L-1 and 1.24x10-5 mol L-1 for HQ, respectively. The novel electroanalytical method was validated in terms of specificity, linearity, range, accuracy and precision (repeatability and intermediate precision) and successfully applied to the direct simultaneous determination of these bleaching agents in real samples. / O trabalho descreve o desenvolvimento e a validação de métodos analíticos para quantificação de duas associações, os antibióticos: sulfametoxazol (SMX) e trimetoprima (TMP); e os agentes despigmentantes: ácido kójico (AK) e hidroquinona (HQ), empregando-se eletrodo de carbono vítreo (ECV) não modificado e voltametria de onda quadrada (VOQ). Para a primeira associação, dois picos de oxidação irreversíveis em 0,96V (SMX) e 1,12V (TMP) foram observados em tampão Britton–Robinson (pH 6,0), caracterizados por processos difusional e adsortivo, respectivamente. Os parâmetros instrumentais da VOQ foram otimizados por meio de um planejamento fatorial completo 23, sendo a melhor resposta em termos de sensibilidade e seletividade obtida em a=30 mV, f=100 s-1 e ΔES=5mV. Nas condições otimizadas, foram construídas curvas analíticas nos intervalos de concentração de 5,5x10-5 a 3,95x10-4 mol L-1 (R=0,9971) para SMX e 1,05x10-5 a 1,04x10-4 mol L-1 (R=0,9974) para TMP. Os valores de limites de detecção (LD) e de quantificação (LQ) calculados foram: 8,52x10-6 mol L-1 e 2,84x10-5 mol L-1 para SMX e 9,31x10-7 mol L-1 e 3,10x10-6 mol L-1 para TMP, respectivamente. O método proposto foi aplicado para determinação simultânea dos fármacos em diferentes amostras farmacêuticas (comprimidos, injeção e suspensão oral) sem a necessidade de nenhum tipo de pré-tratamento da amostra. Os resultados obtidos através do método voltamétrico foram comparados aos obtidos pelo método farmacopeico (HPLC) e comprovou-se que não há diferença significativa em um nível de confiança de 95%. Para a segunda associação, nos voltamogramas cíclicos em tampão Mcllvaine (pH=8,0) um pico de oxidação irreversível em +0,7V foi observado para o AK, enquanto que para HQ foi observado um processo quase-reversível, com picos de oxidação em 0,08V e redução em -0,02V, caracterizados por processos adsortivo e difusional, respectivamente. Utilizando-se os parâmetros da VOQ otimizados (a= 40 mV, f= 40 s-1 e ΔES= 1mV), através de um planejamento fatorial completo 23, as curvas de calibração foram obtidas nas faixas de concentração de 3,5×10-5 a 2,5×10-4 mol L-1 (R= 0,9996) para AK e 1,5×10-5 a 1,6×10-4 mol L-1 (R= 0,9994) para HQ. Os limites de detecção e quantificação foram: 7,84x10-6 mol L-1 e 2,61x10-5 mol L-1 para AK; e 3,71x10-6 mol L-1 e 1,24x10-5 mol L-1 para HQ, respectivamente. O método voltamétrico desenvolvido foi validado em termos de especificidade, linearidade, faixa de concentração, exatidão e precisão (repetibilidade e precisão intermediária) e aplicado na determinação simultânea dos agentes despigmentantes em amostras de clareadores.