published_or_final_version / Industrial and Manufacturing Systems Engineering / Doctoral / Doctor of Philosophy

Thesis (DTech (Chemistry))--Cape Peninsula University of Technology, 2005 / A 2-factorial experiment, involving 3 levels of phosphorus (0, 40, and 80 kg P.ha-I ) as main treatment and 4 cropping systems (mono crop, maize/cowpea inter-row, maize/cowpea intra-row, and maize/cowpea intra-hole cropping) as sub-treatment was conducted in the field for 2 consecutive years in 2003 and 2004 to assess i) the effects of exogenous P supply and cropping system on the concentrations of plant-available nutrients in the rhizosphere of cowpea and maize; ii) the effect of exogenous P supply on tissue concentrations of minerals in nodulated cowpea and maize in mixed plant cultures iii) the effects of exogenous P supply and cropping system on plant growth and N2 fixation, and iv) the effects of exogenous P supply and cropping system on phosphatase activity and microbial biomass in the rhizosphere of cowpea and maize. At harvest, it was found that applying 40 or 80 kg P.ha-I significantly increased cowpea grain yields by 59-65% in 2003 and 44-55% in 2004. With maize, the increases in grain yield were 2037% in 2003 and 48-55% in 2004 relative to zero-P control. In both cropping seasons, the number of pod-bearing peduncles per plant, the number of pods per plant, the number of seeds per pod, and seed yield per cowpea plant were significantly increased with the application of exogenous P. In contrast, these parameters were all significantly depressed by mixed culture relative to mono crop cowpea. Intercropping maize with cowpea produced higher total yields per unit land area than the sole crop counterpart.

Thesis (MTech (Electrical Engineering))--Peninsula Technikon, Cape Town, 2002 / Fermentation is the process that results in the formation of alcohol or organic acids on the basis of growth of bacteria, moulds or fungi on different nutritional media (Ahmed et al., 1982). Fermentation process have three modes of operation i.e. batch, fed-batch and continuous mode ofoperation. The process that interests a lot of control engineers is the fed-batch fe=entation process (Johnson, 1989). The Fed-batch process for the production ofyeast is considered in the study. The considered yeast in the study is the Saccharomyces cerevisiae. It grows in both aerobic and anaerobic environmental conditions with maximum product in the aerobic conditions, also at high concentration of glucose (Njodzi, 2001). Complexity of fedbatch fe=entation process, non-linearity, time varying characteristics, application of conventional analogue controllers provides poor control due to problems in tuning individual loops and the process characteristics. The problem for control of the fedbatch process for the production of yeast is further complicated by the lack of on-line sensors, lack of adequate models as a result of poorly understood dynamics. The lack of on-line sensors results in the impossibility of tuning the analogue controllers in real time. The process for propagation of yeast in aerobic conditions is considered in the dissertation. The experiments are conducted at the University of Cape Town (VCT), Department of Chemical Engineering with a bioreactor and bio-controller are combined in a Biostat ® C lab scale plant (B. Braun Biotech International, 1996). The bio-controller has built in PID controller loops for control variables, with the ability to adjust the controller parameters i.e. P, D and I through the serial interface (Seidler, 1996).

Thesis (MTech (Electrical Engineering))--Cape Peninsula University of Technology, 2005 / Fennentation is the process that results in the fonnation ofalcohol or organic acids on the basis of growth of bacteria, moulds or fungi on different nutritional media (Ahmed et al., 1982). Fennentation process have three modes of operation i.e. batch, fed-batch and continuous ones. The process that interests a lot of control engineers is the fed-batch fennentation process (Johnson, 1989). The Fed-batch process for the production ofyeast is considered in the study. The fennentation is based on the Saccharomyces cerevisiae yeast. It grows in both aerobic and anaerobic environmental conditions with maximum product in the aerobic conditions, also at high concentration of glucose (Njodzi, 200I). Complexity of fed-batch fennentation process, non-linearity, time varying characteristics, application of conventional analogue controllers provides poor control due to problems in tuning individual loops and the process characteristics. The problem for control of the fed-batch process for the production of yeast is further complicated by the lack of on-line sensors, lack ofadequate models as a result ofpoorly understood dynamics. The lack of on-line sensors results in the impossibility oftuning the analogue controllers in real time. The process for propagation of yeast in aerobic conditions is considered in the dissertation. The experiments are conducted at the University of Cape Town (DCT), Department of Chemical Engineering with a bioreactor and bio-controller combined in a Biostat ® C lab scale plant (H. Braun Biotech International, 1996). The bio-controller has built in Pill controller loops for control variables, with the ability to adjust the controller parameters i.e. P, D and I through the serial interface (SeidIer, 1996).

Submitted in fulfillment ofthe requirement for Masters degree oftechnology (Mtech): Electrical engineering, 2002 / Fermentation is the process that results in the formation of alcohol or organic acids on the basis of growth of bacteria, moulds or fungi on different nutritional media (Ahmed et al., 1982). Fermentation process have three modes of operation i.e. batch, fed-batch and continuous mode ofoperation. The process that interests a lot of control engineers is the fed-batch fe=entation process (Johnson, 1989). The Fed-batch process for the production ofyeast is considered in the study. The considered yeast in the study is the Saccharomyces cerevisiae. It grows in both aerobic and anaerobic environmental conditions with maximum product in the aerobic conditions, also at high concentration of glucose (Njodzi, 2001). Complexity of fedbatch fe=entation process, non-linearity, time varying characteristics, application of conventional analogue controllers provides poor control due to problems in tuning individual loops and the process characteristics. The problem for control of the fedbatch process for the production of yeast is further complicated by the lack of on-line sensors, lack of adequate models as a result of poorly understood dynamics. The lack of on-line sensors results in the impossibility of tuning the analogue controllers in real time. The process for propagation of yeast in aerobic conditions is considered in the dissertation. The experiments are conducted at the University of Cape Town (VCT), Department of Chemical Engineering with a bioreactor and bio-controller are combined in a Biostat ® C lab scale plant (B. Braun Biotech International, 1996). The bio-controller has built in PID controller loops for control variables, with the ability to adjust the controller parameters i.e. P, D and I through the serial interface (Seidler, 1996). Even though the used lab scale bio-controller has the ability to monitor certain variables, the automation of the industrial bioreactors is still developing slowly (Dochan and Bastin, 1990) with major problems experienced in modelling and measuring important control variables on-line. This existing situation is due to the characteristics of the fermentation processes as an object of control with highly non-linear, non-stationary, slow dynamics and complex relationships between variables. The existing control strategies on industry are based only of local PID control of some easy for measuring variables. No computer systems for monitoring and optimisation of the process (Morari and Stephanopoulos, 1980). The dissertation is overcoming the mentioned above drawbacks by developing methods, algorithms and programmes for building of a two layer system for optimal control of the Biostat ® C pilot plant with the following subsystems: ~ Data acquisition, ~ Modelling and simulation, ~ Model parameter estimation, ~ Process optimisation, ~ PlO controller parameter tuning, ~ Real time control implementation The system is based on LabVIEW™ and serial communication protocol. The interface between the Bio-start and the host computer is through a standard communication serial port. The development in the dissertation are described as follows: Chapter 1 describes the necessity of the research discussed in the dissertation and highlights comparison between the different approaches for modelling and control of fed-batch processes for the production of yeast, (Johansson, 1993). The aim and the objectives ofthe dissertation are stated and explained. Chapter 2 describes the process as an object of control looking precisely at the influence of the physiochemical variables on the biological variables. The relationship is identified through the enzymes. The results from previous experiments are discussed to illustrate the constraints associated with the control of the process under study Chapter 3 describes the different types of models as applicable to the dissertation. The comparison between different types is highlighted. The derivation of the developed yeast model using mass balance equations and rate laws is discussed and presented in the chapter. The problem for simulation of the model is solved using Matlab and LabVIEW™ programs. Chapter 4 describes the formulation of the problem for estimation of the fed-batch model coefficients and the method, algorithm and programme developed to solve this problem. In chapter 5 the optimal control problem is formulated and solved using optimal control theory, the approach of the functional of Lagrange is used. The optimisation layer problems are determined and based on the solutions of the previous upper layer i.e. the model parameters from the adaptation layer. The optimal operation of the process or yield of the yeast is based on some criteria for the production of biomass, and some constraints over minimal and maximal values of the variables. Decomposition method to solve the optimal control problem is developed on the bases of an augmented functional of Lagrange and decomposition in time domain. Algorithm of the method and program in Matlab are developed. Tuning of PID parameters for the controllers in Biostat ® C is described based on the optimal trajectories for the physiochemical variables obtained from the optimal control problem solution. Chapter 6 presents algorithms and programmes for monitoring and real time control of fed-batch fermentation process for the production of yeast, using Personal Computer, B Braun Biostat ® C Lab scale fermentation unit and LabVIEW™ as driver software. Hardware and software parts of the control systems are described and discussed. LabVIEW™ code is described. Chapter 7 presents the users manual. The mam functionality of the developed application and programmes is described and discussed. The source code ofthe developed programmes is presented in chapter 8. Chapter 9 presents the conclusion highlighting the developments in the dissertation as well as the future work on the topic and the possible application of the developed work in industry on the bigger scale fermentors. The positive characteristics of the developed methods algorithms and programmes are: -7 The developed model incorporates the physiochemical variables in the biological mass balance equations. In this way: the influence of the enzymes over the biological variables is utilised and possibilities for process optimisation is created. The process can be optimised m both physiochemical and biological variables. The physiochemical variables can be used as control inputs to reach the process optimisation. Data acquisition system gIves good possibilities for analysis and simulation ofthe process. The optimal control of the process is achieved without using expensive online sensors for measurement of the biological variables. This is why the developed system is applicable to the existing hardware and software control and measurement systems in industry. The control system automates the operation of the lab scale fermentation unit. It is safe, stable and operational.

This study utilizes a Monte Carlo simulation to examine the performance of multivariate geometric moving average control chart schemes for controlling the mean of a multivariate normal process. The study compares the performance of the proposed method with a multivariate Shewhart chart, a multiple univariate cumulative sum (CUSUM) control chart, a multivariate CUSUM control chart and a multiple univariate geometric moving average control chart.

No description available.

The semiconductor industry is highly competitive environment where modest improvements in the manufacturing process can translate to significant cost savings. An area where improvements can be realized is reducing the number of wafers that fail to meet their electrical specifications. Wafers that fail to meet electrical specifications are scrapped, which negatively impacts yield and increases manufacturing costs. Most of the existing semiconductor process control research has focused on controlling individual steps during the manufacturing process via run-to-run control, but almost no work has looked at directly controlling device electrical characteristics. Since meeting electrical specifications is so critical to reducing scrap a fab-wide electrical parameter control scheme is proposed to directly control electrical parameter values. The goal of the controller is reducing the variation in the electrical parameters. The control algorithm uses a model to predict electrical parameter values after each processing step. Based on this prediction the decision to make a control move is made. If a control move is necessary, optimal adjustments for the subsequent processing steps are determined. The process model is continually updated so that it reflects the current process. A simple implementation using a least squares model is first proposed. Simulations and an industrial case study demonstrate the potential improvements that can be achieved with the algorithm and the limitations of the simple implementation are discussed. A partial least squares modeling and control algorithm combined with missing data algorithms are proposed as enhancements to the electrical parameter control algorithm to address many of the issues faced when implementing such a control strategy in real manufacturing environments. The enhancements take the input variable correlations into account when making control moves and utilize the correlation structure to make better model predictions. Simulations are performed to determine the effectiveness of the enhancements. A cost function formulation and a Bayesian based alternative are also presented and evaluated. The cost function implementation uses a different method to determine the optimal set points for the subsequent processing steps than the other implementations use. Simulations are used to compare the cost function formulation with the other methods presented. The Bayesian implementation addresses the stochastic nature of the manufacturing process by dealing with the probabilities of events occurring. A simulation of the Bayesian algorithm is preformed and the algorithms limitations are discussed.

This dissertation is composed of three major parts, each studying a problem related to semiconductor manufacturing. The first part of the dissertation proposes a high-level scheduling model that serves as an intermediate stage between planning and detailed scheduling in the usual planning hierarchy. The high-level scheduling model explicitly controls the WIP over time in the system and provides a more specific guide to detailed scheduling. WIP control is used to balance the WIP (Work In Process) level and to keep the bottleneck station busy to maintain a high throughput rate. A mini-fab simulation model is used to evaluate the benefits of different approaches to implementing such a high-level scheduling model, and to compare different WIP control policies. Extensive numerical studies show that the proposed approaches can achieve much shorter cycle times than the traditional planning-scheduling approach, with only a small increase in inventory and backorder costs. With increasing worldwide competition, high technology product manufacturing companies have to pay great attention to lower their production costs and guarantee high quality at the same time. Advanced process control (APC) is widely used in semiconductor manufacturing to adjust machine parameters so as to achieve satisfactory product quality. The interaction between scheduling and APC motivates the second part of this dissertation. First, a single-machine makespan problem with APC constraints is proved to be NPcomplete. For some special cases, an optimal solution is obtained analytically. In more general cases, the structure of optimal solutions is explored. An efficient heuristic algorithm based on these structural results is proposed and compared to an integer programming approach. Another important issue in manufacturing system is maintenance, which affects cycle time and yield management. Although there is extensive literature regarding maintenance policies, the analysis in most papers is restricted to conventional preventive maintenance (PM) policies, i.e., calendar-based or jobbased PM policies. With the rapid development of new technology, predictive maintenance has become more feasible, and has attracted more and more attention from semiconductor manufacturing companies in recent years. Thus, the third problem considered in this dissertation is predictive maintenance in an M/G/1 queueing environment. One-recipe and two-recipe problems are studied through semi-Markov decision processes (SMDP), and structural properties are obtained. Discounted SMDP problems are solved by linear programming and expected machine availabilities are calculated to evaluate different PM policies. The optimal policy can maintain a high machine availability with low long-run cost. The structures of the optimal PM policies show that it is necessary to consider multiple recipes explicitly in predictive maintenance models. / text