This work is aimed at understanding the thermodynamic principles influencing the phosphorus partition between slag and steel during steelmaking, particularly during refining of high-phosphorus hot metal using the basic oxygen steelmaking (BOS) process. Mapping of the slag path has been carried out using a mass balance model based on input conditions and off-gas analysis, which has been validated by intermediate measurements of slag and metal composition in an industrial basic oxygen furnace (BOF). The slag composition is found to vary over a wide range of basicity (%CaO/%SiO2) and iron oxide content. The partition of phosphorus increases with progress of refining but reduces during the intermediate period, corresponding to a decrease in iron oxide concentration and formation of “dry slag”. The phosphorus partition ratio increases again towards the end of the process. The equilibrium partition ratios for such conditions have been theoretically estimated using slag “model(s)”, based on the estimated slag compositions. It is seen that the partition ratio of phosphorus remains within 45% – 60% of the equilibrium value, showing a gradual decrease with progress of the refining process. Equilibrium phosphorus partition ratios for slags containing low concentrations of MnO and Al2O3 have been experimentally determined, over the ranges of basicity and iron oxide concentration approximately corresponding to that observed in the typical BOS process. It is seen that the equilibrium phosphorus partition ratio is practically independent of basicity greater than 2.5 – 2.6, over the entire range of temperature and FeOx concentration studied. Variation of the activity of P2O5 with basicity and FeOx concentration has been investigated to explain the trends observed in the variation of phosphorus partition ratio. It is seen that the activity coefficient of P2O5 is lowest, and hence most conducive for removal of phosphorus from steel, over certain finite ranges of basicity and iron oxide concentration. It follows that the efficiency of dephosphorisation is likely to be maximum if the basic oxygen steelmaking process is operated within these ranges. Simultaneous with this work, mathematical correlations have been developed for estimation of P2O5 activity coefficient and phosphorus partition ratio as functions of slag composition. The correlations are compared with those proposed by earlier workers and are found to result in better predictions over certain composition ranges. / QC 20100623

In the present work, viscosities and sulphide capacities of high alumina blast furnace slags were investigated. The systems investigated were four component CaO-SiO2-MgO-Al2O3 quaternary system, CaO-SiO2-MgO-Al2O3-TiO2 and CaO-SiO2-MgO-Al2O3-CaF2 quinary systems. Viscosities of high alumina blast furnace slags were experimentally determined by the rotating cylinder method using Brookfield digital viscometer model LVDV-II+ pro. Experiments were conducted in the temperature range of 1573- 1873 K. The effects of temperature, basicity, TiO2, CaF2 and silica activity of slags on viscosity were studied. Viscosity decreases with basicity for high alumina blast furnace slags with increase in basicity and CaF2. At higher basicity (~0.8), slag viscosity decreases even with small amount of TiO2 (~2%) addition in the slag. With increase in silica activity in the range of 0.1 to 0.4,viscosity of slag increases and the increase is steeper below liquidus temperature. Sulphide capacity of the slag was measured using gas-slag equlibria. The liquid slag was equilibrated with Ar-CO-CO2-SO2 gas mixture. The slag systems studied were the same as in the case of viscosity measurements. Experiments were conducted in the temperature range of 1773 to 1873 K. Effect of temperature, basicity, MgO,TiO2 and CaF2 contents of slags on sulphide capacity were studied. As expected, sulphide capacity was found to increase with increase in temperature and basicity. At higher experimental temperature (~ 1873 K) TiO2 was found to decrease the sulphide capacity of slags. But, at lower temperature, there was no significant effect of TiO2 on the sulphide capacity. Sulphide capacity increases with increase in MgO content of slag if MgO content is more than 5%. Based on above experimental data, models were developed for estimation of viscosity and sulphide capacity of blast furnace slags. These models were later on applied for designing the slags for achieving the optimum slag characteristics so that slag volume can be reduced. With the help of these models slag volume was reduced to the extent of 5-10 kg per ton hot metal and also silicon content of the hot metal was reduced by around 10% with some improvement in slag viscosity and sulphide capacity of the slag. / QC 20100818

During the production of steel the oxide inclusion content partly depends on the reaction of the melt with the furnace lining, the ladle lining and the pouring system. The refractory material may be eroded by the molten steel and slag as well as corroded through chemical reactions with the slag and molten steel and the deoxidation products. In this report the effects of revolution speed, temperature and steel composition on the rate of dissolution of commercial MgO-C refractory samples into Al-deoxidised molten steel and CaOAl2O3- SiO2-MgO slag were examined by the rotating cylinder method. The study also includes tests with slag were doloma refractory samples are examined by the same method. Cylinders of MgO-C refractory material were immersed in to steel that was deoxidised by adding metallic aluminium. This was carried out in the temperature range of 1873 to 1973C° and at rotational speeds of 100 to 800 rpm for different holding times. The experimental results show that the rate of dissolution of MgO-C refractory materials increased with the temperature, rotational speed and immersion time. This supports the assumption that the diffusion of magnesium through the slag boundary layer formed around the refractory samples would be the rate-determining step. Mass transfer coefficients calculated on the basis of experimental results are in good agreement with earlier published results for pure ceramics. A formation of a thin oxide layer at the interface was found. It is due the reaction between magnesium vapour and the CO generated by the reaction MgO and C in the refractory walls. The oxide inclusions formed in the steel have been shown to mainly consist of MgO, Al2O3 and a mixture of them. The rate of dissolution of solid MgO-C into liquid CaO-Al2O3-SiO2-MgO slag at different temperatures was studied under conditions of forced convection by rotating cylindrical refractory specimens in a stationary crucible containing the molten slag similar to the MgO-C refractory/steel experiments. The corrosion rate was calculated from the change in diameter of the cylindrical specimens. The specimens were rotated for 15 to 120 minutes at speeds of 100 to 400 rpm in the molten slag. The rate of corrosion increased with temperature and with rotating speed of the rod and decreased when the slag was nearly saturated with MgO. The experimental results confirm the assumption that the diffusion of magnesium oxide through the slag phase boundary layer controls the corrosion process. The corrosion mechanism seems to be the dissolution of elements in the refractory materials into the slag, followed by penetration into the pores and grain boundaries. Finally, grains are loosened from the refractory into the slag. The investigation of doloma and doloma-carbon showed that the dissolution of magnesia into the slag was determining the corrosion rate. As for the other experiments, steel/MgO-C refractory and slag/MgO-C refractory, the corrosion rate was calculated from the change in diameter of the cylindrical specimens. The specimens were rotated for 15 to 120 minutes at speeds of 100 to 400 rpm in the molten slag. The results from the study showed that refractory materials that were impregnated with carbon had a much better slag resistance than the refractory that contained no carbon. This is due to the higher wetting angle between carbon and slag. Corrosion of MgO-C refractories in different gas atmospheres consisting of air, Ar, CO or Ar/CO was also studied. Experiments were carried out in the temperature range 1173 K to 1773 K and for holding times between 2 to 120 min. The reaction rate of the MgO-C material was determined from measurements of the weight loss of the samples. The results showed that the refractory weight loss increased with an increased temperature or an increased holding time. The thermodynamic conditions and the experimental results show that magnesium gas and carbon monoxide gas should form during ladle refining of steel when the refractory material consists of MgO-C. / QC 20100813

Today’s problems with emissions of green house gases, land filling of waste and depletion of the oil reserves calls for new energy systems based on alternative fuels like biomass and waste. Gasification is an attractive technology for the use of such solid fuels. Conventional gasification, in the vast majority of cases, uses in-reactor heat release from combustion of part of the feedstock, possibly coupled with a limited preheating of the agent, to obtain the necessary temperatures in the gasifier bed. During recent years, a new gasification technology, using highly preheated gasification agents (> 1273 K), has been developed. The extra heat brought into the process by the high temperature agent reduces the amount of feedstock that has to be oxidized to supply the necessary heat and the use of highly preheated agents has previously proven to have several positive effects on the fuel gas quality.In difference to the previous work on gasification with highly preheated agents, this thesis primarily focuses on the fundamental aspects namely, mass conversion, heating and ignition. It starts by considering single fuel particles or thin beds of fuel particles inserted into highly preheated agents. Mass conversion, heating and ignition are reported in function of the temperature and oxygen concentration of the agent and formulas for the prediction of ignition time and ignition mechanism are developed. The perspective is then widened to include the whole gasifier bed. Simulations of fixed bed batch gasification using highly preheated agents are performed with a mathematical model and used to study how the high agent temperature influences the mass conversion, devolatilisation front rate and the temperature distribution in the fixed fuel bed. Further, the gas quality and gasification efficiency are studied by means of large scale experiment. Ultimately, a thermodynamic analysis of the whole autothermal gasification system, including both a regenerative preheating system and the gasifier, is made.The particle study reports results from experiments with wood and coal and agents consisting of mixtures of nitrogen and oxygen in various proportions. It is shown that an increase in agent temperature from 873 K to 1273 K make the conversion process faster, mostly due to an early onset of the devolatilisation (fast drying) but also due to an increased devolatilisation rate (at least in the case of wood). The time to ignition also decreases significantly, particularly so between 873 and 1073 K. Further, it is shown that the higher the agent temperature, the more pronounced was also the tendency of the coal particles to heat significantly faster in oxygen diluted conditions (5,10 and 21% oxygen) than in inert (0% oxygen) or oxygen rich conditions (30, 50, 80 and 100% oxygen). An increase in agent temperature is also shown to reduce the dependency of the process on the oxygen concentration, at least in diluted conditions (5-21% oxygen). The results also indicate that for coal an increase in the oxygen concentration, specifically in the region above the atmospheric concentration, leads to a decreased dependency on the agent temperature. It is finally shown in the experiments with agent temperatures of 1073 and 1273 K that a flame is promptly formed even in very low concentrations of oxygen.The gasifier study reports results from simulation of batch air gasification and experiments in both batch and continuous up-draft fixed bed gasifier with wood and waste derived fuel and air and mixtures of air and steam. It is shown that the conversion process is faster the higher the air temperature. In particular somewhere between air temperatures of 623 K and 803 K the process behaviour changes. In fact, the devolatilisation rate is significantly increased in this region while it increases less sharply with air temperature below and above this temperature window. The temperature distribution in the bed shows less sharp gradients at high temperature (> 803 K) than at low temperatures (< 623 K). It is also showed experimentally and in fairly large scale that the use of highly preheated air for the gasification of biomass and waste derived fuels can produce - in continuous mode – relatively high yields of product syngas with relatively high fractions of combustible gases and probably also low content of tar. The efficiency of the gasification under these conditions, even when the extra heat input in the preheated agent is considered in the computation of the gasification efficiency, is shown to be comparable to that of conventional gasification techniques. The results also shows that with the use of steam in the agent, the content of hydrogen can be further increased with respect to gasification with only preheated air.In base of the results of the particle study and the gasifier study it is shown that a there exists two regimes of operation in function of the agent temperature, separated by the minimum agent temperature to guarantee spontaneous ignition regardless of the particle temperature. The value of this temperature depend on material properties and the kinetics of the reaction, thus also on the oxygen concentration. When agent temperatures below the minimum agent temperature to guarantee spontaneous ignition regardless of the particle temperature are used, the drying and devolatilisation are mainly controlled by the heat released by reactions. The heating of the fuel particles and their devolatilisation are relatively slow and the devolatilisation rate is highly oxygen dependent. In a fixed bed, the devolatilisation front rate is low and the bed is characterised by significant temperature gradients.When the agent temperature is higher than the minimum agent temperature to guarantee spontaneous ignition regardless of the particle temperature, the drying and devolatilisation are mainly controlled by the convective heat transfer from the preheated agent and the released volatiles ignite very fast even in diluted conditions. This results in very efficient heat transfer to the fuel particles. In the fixed fuel bed the process is characterized by a high devolatilisation front rate. Thus, the temperature gradients in the bed are significantly reduced and the gasification can be said to be thermally homogeneous. Thanks to high rates of heat transfer and mass conversion, the heating value of the dry produced syngas is high with high concentrations of combustible species. The ignition of the volatiles and the high temperatures all along the bed presumably contributes to the reduction of the tar content even in up-draft configurations. The high temperatures also allows for operation with reduced air – to – fuel ratios which further increased the value of the produced gas (thanks to less dilution by nitrogen).The system study presents a concept for an autothermal system including both preheating and gasification. Results from a thermodynamic analysis of such a system are reported. Autothermal operation of a thermally homogeneous gasifier is possible only in a twin component system in which the gasifier is coupled to a preheating system able to reach preheating temperatures well above the minimum agent temperature to guarantee spontaneous ignition regardless of the particle temperature. It is shown that to reach certain temperature levels of the gasification air, heat exchange between product gas and air is not enough and the preheating system has to improve the temperatures involved, for example by burning part of the produced gas in a regenerative preheater. Further, it is shown that in comparison to gasifier without such a system for additional preheating, the autothermal Thermally Homogeneous Gasification system has the ability to significantly improve the gas quality (in terms of heating value of the dry gas) without losing energy- or exergy efficiency to an appreciable extent. / QC 20100903

<p>The topic of this thesis is estimation of the dynamic changes of the surface temperatureand heat flux during heating processes by using an inverse method. The local transient surface temperature and heat flux of a steel slab are calculated based on measurements in the interior of the slab.</p><p>The motivations for using an inverse method may be manifold. Sometimes, especially in the field of thermal engineering, one wants to calculate the transient temperature or heat flux on the surface of a body. This body may be a slab, or billet in metallurgical applications. However, it may be the case that the surface for some reason is inaccessible to exterior measurements with the aid of some measurement device. Such a device could be a thermocouple if contact with the surface in question is possible or a pyrometer if an invasive method is preferred. Sometimes though, these kinds of devices may be an inappropriate choice. It could be the case that the installation of any such device may disturb the experiment in some way or that the environment is chemically destructive or just that the instruments might give incorrect results. In these situations one is directed to using an inverse method based on interior measurements in the body, and in which the desired temperature is calculated by a numerical procedure.</p><p>The mathematical model used was applied to experimental data from a small scale laboratory furnace as well as from a full scale industrial reheating furnace and the results verified that the method can be successfully applied to high temperature thermal applications.</p>

<p>In this master thesis, optical dissolved oxygen (DO) sensors are compared to standard electrochemical DO-sensors. The optical sensors are also compared against each other. DO is an important parameter to measure and to control during microorganism fermentation.  A DO-sensor needs to fulfill certain criteria, such as; purposive response time, the ability to show accurate and stable readings without drift over time, have the capability to manage several sterilizations and exhibiting membrane fouling resistance. The reference used to test these parameters was an industrially accepted electrochemical sensor. By comparing the two technologies, optical sensors showed similar abilities to perform regarding response time and drift performance during the longer test runs. But the optical sensors showed a better performance during sterilizations, with less drift in raw value. When using electrochemical sensors as a reference, different brands of optical sensor gave different results when testing for accuracy at different concentration levels. Due to error in the method, tests for fouling were inconclusive. However, membranes from the two kinds of technologies showed an equal hydrophobic character. The two participating optical DO sensors, Mettler Toledo InPro 6990i and Hamilton Visiferm were compared regarding ability to manage sterilizations and generate accurate measuring values at different DO concentrations. The two sensor brands showed an equal behavior during the fifteen test sterilizations. The Mettler Toledo InPro 6990 had better abilities to follow the linear behavior of the electrochemical reference sensor. While the  Hamilton Visiferm sensor differed at most 3.5% units from the electrochemical reference values. Hamilton Visiferm was tested with analogue signal, which gave a noise problem. When using digital signal the noise disappeared. The goal of this study was to recommend a sensor, for Belach Bioteknik AB, and the recommendation would be Hamilton Visiferm, for being practical as well as reasonable priced without renounce of measuring quality.</p>

In this master thesis, optical dissolved oxygen (DO) sensors are compared to standard electrochemical DO-sensors. The optical sensors are also compared against each other. DO is an important parameter to measure and to control during microorganism fermentation.  A DO-sensor needs to fulfill certain criteria, such as; purposive response time, the ability to show accurate and stable readings without drift over time, have the capability to manage several sterilizations and exhibiting membrane fouling resistance. The reference used to test these parameters was an industrially accepted electrochemical sensor. By comparing the two technologies, optical sensors showed similar abilities to perform regarding response time and drift performance during the longer test runs. But the optical sensors showed a better performance during sterilizations, with less drift in raw value. When using electrochemical sensors as a reference, different brands of optical sensor gave different results when testing for accuracy at different concentration levels. Due to error in the method, tests for fouling were inconclusive. However, membranes from the two kinds of technologies showed an equal hydrophobic character. The two participating optical DO sensors, Mettler Toledo InPro 6990i and Hamilton Visiferm were compared regarding ability to manage sterilizations and generate accurate measuring values at different DO concentrations. The two sensor brands showed an equal behavior during the fifteen test sterilizations. The Mettler Toledo InPro 6990 had better abilities to follow the linear behavior of the electrochemical reference sensor. While the  Hamilton Visiferm sensor differed at most 3.5% units from the electrochemical reference values. Hamilton Visiferm was tested with analogue signal, which gave a noise problem. When using digital signal the noise disappeared. The goal of this study was to recommend a sensor, for Belach Bioteknik AB, and the recommendation would be Hamilton Visiferm, for being practical as well as reasonable priced without renounce of measuring quality.

The liquid steel sampling method is one of the commonly used procedures in monitoring the steelmaking process. Besides it can be used for analyzing the dissolved alloys, hydrogen content and oxygen content, it can be also employed to monitor the inclusion characteristics at the steelmakings. Here, a crucial point is that the steel sampler should be filled and the metal solidifies without changing the inclusion characteristics. Therefore, the objective of this work is to fundamentally understand the liquid steel sampling process by means of analyzing and modeling the two-phase flow during the sampler filling process, and verifying the mathematical model by using the experimental data. The present dissertation presents an experimental and theoretical study of the filling process of both the lollipop-shaped sampler and the rectangular-shaped sampler. Firstly, a physical modeling by using a water model has been carried out to fundamentally investigate the flow pattern inside the sampler vessels during its filling. The flow patterns were obtained by a PIV system. Then, a mathematical model has been built to theoretically understand the phenomena. The commercial CFD code was used. Here, different turbulence model have been compared between the realizable k-ε turbulence model and Wilcox k-ω turbulence model. It concludes that the Wilcox k-ω turbulence model agrees well with the PIV measurements.HH Thus, the preferred it was further employed to predict the turbulent flow inside the production lollipop-shaped sampler fillings. It is important to find that the average collision volume in the production steel sampler without solidification at filling is about 30 times higher than that in a ladle furnace. In the end, the whole sampling system was modeled. The initial solidification during the filling was taken into account. Focus was on the influence of the initial solidification on the inclusion concentrations. A discrete phase model was used to simulate the movement of inclusions in the liquid steel. Some selected different sized primary inclusions that exist in the ladles at a steelmaking process were simulated. The same method of studying the filling procedure of the lollipop-shaped sampler was further applied to comprehensively investigate the rectangular-shaped sampler. / QC 20100908

The goals of this Bachelors’ Thesis were to establish the production loss through the VTIS1 Pasteur at Arla Foods Linköping dairy and then suggest simple but effective solutions to minimize the losses for each production run. If the timeframe allowed it, some of the solutions would be implemented and tested in full production to determine their impact on economic and environmental savings. The very first step of this project was to establish the statistics over what losses there is in the production pipeline through VTIS1, from milk and powders in mixer tanks to finished and packed products in cold storage. Then, when knowing how much product was actually lost and where, those bottlenecks could be pinpointed and examined more closely. At this stage the focus was switched, from the otherwise complicated and broad manufacturing process, to the VTIS1-aggregate, since early data indicated there were quite substantial losses there. In conjunction with this project Arla Foods purchased an absorption photometer-measuring device (OPTEK AS16) that could be inserted straight into the product stream out of VTIS1, allowing real time measurements of its optical density. A substantial series of measurements of the product stream (through all of November) indicated that a certain amount of pure quality product was discarded into the factory sewers with each production batch. Volumes of the product losses clearly depended on which sterile tank it was sent to, due to the different lengths of the pipelines. Through the production year of 2007 the VTIS1 aggregate pasteurised over 1500 batches of various products and all preliminary data indicates this number to increase significantly in 2008. Since most products going through VTIS1 contain both milk/cream and some flavourings or jams in various compositions, this means they are a fair bit more expensive per litre in comparison to plain milk. Since the VTIS1 aggregate is positioned fairly early in the production line it was deemed logical to start there to eliminate losses and then, if time would allow, move further down the stream. The time assigned for the project did allow for both creating a statistics database over product losses and the implementations of a few suggested adjustments. The report includes the estimated financial savings of this project for 2008; the numbers are based on how much the adjustments would have saved Arla Foods in economic losses through 2007. This includes the costs for the pure volumes of products as well as the estimated savings of environmental costs for waste treatment during that same period, based on the products’ COD, S-tot, N-tot and Ts contents. / Examensarbetets syfte har bestått i att med statistik kartlägga produktförlusterna i genom Arla Foods i Linköpings VTIS1 produktion och föreslå möjliga lösningar för att minska råvarusvinnet. Detta skall åstadkommas med hjälp av nya smarta lösningar på processoptimeringar, ändringar i körrutiner och bättre produktplanering. Projektbeskrivningen var ursprungligen med avsikt ganska vag, för att sedan efter litet efterforskning fokusera uppmärksamheten på området kring Mixertankarna, VTIS1 och Steriltankarna. I studien har vikten lagts på rent tekniska svinn, som sker vid varje körning och inte mätbart påverkas av den mänskliga faktorn, detta för att få en repeterbarhet och stabila mätvärden. Statistiksammanställning och kartering av de olika produktförlusterna gjordes i Excel, där data hämtades både via Arla Foods AS400 server samt logisticprogrammet Movex Explorer som hanterar lagerförningar, men även via ABB: s WebProInfo server och trendverktyget MES. En snabb analys av den förenklade statistiken visade på att mycket av de förluster som uppstod i tillverkningen genom VTIS1 faktiskt uppstod i själva aggregatet och det var alltså här man beslutade att, i samråd med Timo Kjellberg och Gunnar Jangbrand, begränsa projektet till just mixertankarna, VTIS och steriltankarna. Vid planeringen av examensarbetet var tanken att gå igenom hela tillverkningsprocessen, från det att produkten blandats och behandlats, till det att den förpackats och transporterats till lagerlokalen. Av tidstekniska skäl har alltså tidfördelningen fokuserats till att omfatta en liten men dock betydelsefull del i produktionen. Processen i sin helhet är alltför komplex för att sätta sig in i på såpass kort tid och detta tillvägagångssätt har gett goda resultat, mycket på grund av att den tid som funnits kunnat fokuseras på att noga dokumentera, analysera och förbättra tillverkningsprocessen från första början. Målsättningen för arbetet är att kunna minimera det tekniska svinn som uppkommer i varje ny produktionscykel och på så sätt både spara Arlas pengar och miljön, en faktor som i rapporten har räknats om till en ren ekonomisk besparing i form av minskade miljöavgifter. Fler möjliga förbättringar tas upp senare i rapporten.

<p>The goals of this Bachelors’ Thesis were to establish the production loss through the VTIS1 Pasteur at Arla Foods Linköping dairy and then suggest simple but effective solutions to minimize the losses for each production run. If the timeframe allowed it, some of the solutions would be implemented and tested in full production to determine their impact on economic and environmental savings.</p><p>The very first step of this project was to establish the statistics over what losses there is in the production pipeline through VTIS1, from milk and powders in mixer tanks to finished and packed products in cold storage. Then, when knowing how much product was actually lost and where, those bottlenecks could be pinpointed and examined more closely. At this stage the focus was switched, from the otherwise complicated and broad manufacturing process, to the VTIS1-aggregate, since early data indicated there were quite substantial losses there.</p><p>In conjunction with this project Arla Foods purchased an absorption photometer-measuring device (OPTEK AS16) that could be inserted straight into the product stream out of VTIS1, allowing real time measurements of its optical density. A substantial series of measurements of the product stream (through all of November) indicated that a certain amount of pure quality product was discarded into the factory sewers with each production batch. Volumes of the product losses clearly depended on which sterile tank it was sent to, due to the different lengths of the pipelines.</p><p>Through the production year of 2007 the VTIS1 aggregate pasteurised over 1500 batches of various products and all preliminary data indicates this number to increase significantly in 2008. Since most products going through VTIS1 contain both milk/cream and some flavourings or jams in various compositions, this means they are a fair bit more expensive per litre in comparison to plain milk.</p><p>Since the VTIS1 aggregate is positioned fairly early in the production line it was deemed logical to start there to eliminate losses and then, if time would allow, move further down the stream.</p><p>The time assigned for the project did allow for both creating a statistics database over product losses and the implementations of a few suggested adjustments. The report includes the estimated financial savings of this project for 2008; the numbers are based on how much the adjustments would have saved Arla Foods in economic losses through 2007. This includes the costs for the pure volumes of products as well as the estimated savings of environmental costs for waste treatment during that same period, based on the products’ COD, S-tot, N-tot and Ts contents.</p> / <p>Examensarbetets syfte har bestått i att med statistik kartlägga produktförlusterna i genom Arla Foods i Linköpings VTIS1 produktion och föreslå möjliga lösningar för att minska råvarusvinnet. Detta skall åstadkommas med hjälp av nya smarta lösningar på processoptimeringar, ändringar i körrutiner och bättre produktplanering.</p><p>Projektbeskrivningen var ursprungligen med avsikt ganska vag, för att sedan efter litet efterforskning fokusera uppmärksamheten på området kring Mixertankarna, VTIS1 och Steriltankarna. I studien har vikten lagts på rent tekniska svinn, som sker vid varje körning och inte mätbart påverkas av den mänskliga faktorn, detta för att få en repeterbarhet och stabila mätvärden.</p><p>Statistiksammanställning och kartering av de olika produktförlusterna gjordes i Excel, där data hämtades både via Arla Foods AS400 server samt logisticprogrammet Movex Explorer som hanterar lagerförningar, men även via ABB: s WebProInfo server och trendverktyget MES.</p><p>En snabb analys av den förenklade statistiken visade på att mycket av de förluster som uppstod i tillverkningen genom VTIS1 faktiskt uppstod i själva aggregatet och det var alltså här man beslutade att, i samråd med Timo Kjellberg och Gunnar Jangbrand, begränsa projektet till just mixertankarna, VTIS och steriltankarna.</p><p>Vid planeringen av examensarbetet var tanken att gå igenom hela tillverkningsprocessen, från det att produkten blandats och behandlats, till det att den förpackats och transporterats till lagerlokalen.</p><p>Av tidstekniska skäl har alltså tidfördelningen fokuserats till att omfatta en liten men dock betydelsefull del i produktionen. Processen i sin helhet är alltför komplex för att sätta sig in i på såpass kort tid och detta tillvägagångssätt har gett goda resultat, mycket på grund av att den tid som funnits kunnat fokuseras på att noga dokumentera, analysera och förbättra tillverkningsprocessen från första början.</p><p>Målsättningen för arbetet är att kunna minimera det tekniska svinn som uppkommer i varje ny produktionscykel och på så sätt både spara Arlas pengar och miljön, en faktor som i rapporten har räknats om till en ren ekonomisk besparing i form av minskade miljöavgifter. Fler möjliga förbättringar tas upp senare i rapporten.</p>