3. Electronic Theses and Dissertations (ETDs) - All submissions
Permanent URI for this communityhttps://wiredspace.wits.ac.za/handle/10539/45
Browse
2 results
Search Results
Item Pattern recognition of social contact events from wearable proximity sensor data using principal component analysis(2019-06) Makhasi, Mvuyo KhuseloData from wearable proximity sensors can be used to measure and describe social contact patterns between individuals in a household. Previous work describing contact patterns, has been qualitative and relies on visual, subjective observations. Data of this kind has been collected for a short period of measurement ranging from 2-3 days. An automated, quantitative analysis of contact patterns could enable an accurate and new representation of social contact patterns. Data was collected from ten households, for 21 days in a pilot study implemented in South Africa. 20 datasets were analysed, representing contact events of 20 individuals. Principal Component Analysis was implemented to determine the similarity of contact events across the days of the experiment and to estimate the minimum number of days required to be sampled, to validly represent an individual’s contact activity. The results show that there is a great variation in contact activity across the days of the experiment, as represented by the number of clusters of similar days. The minimum number of days required was determined by the number of days that had a significant contribution to the first three principal components and this varied across individuals from 5 – 11 days. Further analysis on a larger cohort has a potential to provide better social contact parameters for complex social behavioural models and may assist in understanding transmission dynamics of respiratory pathogens, needed in public health research.Item Metaheuristic approaches for scheduling of multipurpose batch plants(2018) Woolway, Mathew JohnThe field of batch chemical process has seen a significant rise in research over the last five decades as changes in the economic climate have lead to an increased demand for the manufacturing of high-value small-volume products. Due to the dependency on time, batch processes are considerably more complex than their continuous process counterparts. The predominant approach in batch process literature makes use of mathematical programming, whereby binary variables are utilised to indicate the assignment of certain tasks to capable units. This mathematical programming strategy, coupled with the aforementioned time complexity can lead to computational intractability due to the extended enumeration of binary variables. In this thesis, the reduction of computational time requiredinthesolutionofmultipurposebatchplantschedulingisconsidered. Due to the infeasible computational times required to solve mathematical programming models in multipurpose batch plant scheduling, often close-to optimal solutions rather than global optimal solutions are accepted. If close-to optimal solutions are acceptable then it is reasonable to explore non-deterministic metaheuristic strategies to reduce the required computational time. In order to apply these strategies, generalised frameworks consistent with metaheuristic approaches are necessary. Presently, no decoupled generalised framework suitable for various metaheuristic implementation exists in the literature. As a result, this thesis presents two novel mathematical frameworks for the representation of batch scheduling. Specifically, one framework for discrete-time approaches and another framework for continuous-time approaches. In each framework, two well-known literature examples are considered. In addition, three metaheuristic techniques are applied to these literature examples, namely, genetic algorithms (GA), simulated annealing (SA) and migrating bird optimisation (MBO). The resultant framework allows for experimentation of 12 variants of the literature examplestobeinvestigated,whichcanbecomparedtothecurrentlyacceptedmixed integerlinearprogramming(MILP)approach. In the aforementioned experiment, simulated results with the metaheuristics implemented under the newly introduced frameworks showed a reduction in computational time of up to 99.96% in the discrete-time approach and 99.68% in the continuous-time approach. Additionally, the genetic algorithm showed to be the best performer of the metaheuristic suite, often obtaining the global optimum in short-time horizons and close-to-optimal solutions in the medium-to-long time horizons. Furthermore, parallel implementations were explored and showed additional time reduction would be possible, with certain workloads terminating 2 ordersofmagnitudelessincomputationaltimethanserialimplementations. Theresultsshowtheapplicationofmetaheuristicstotheschedulingofmultipurpose batch plants are indeed appropriate and are able to obtain close-to-optimal solutions to that of their MILP counterparts at considerably reduced computational times.