Research Articles (Statistics)

Permanent URI for this collectionhttp://hdl.handle.net/2263/1835

A collection containing some of the full text peer-reviewed/ refereed articles published by researchers from
the Department of Statistics

Browse

Now showing 1 - 20 of 527

Behaviour of commercial layer hens in enriched cages and floor systems examined using human and video-based observations
(South African Society for Animal Science, 2025-06) Chilemba, Timothy.; Van Marle-Koster, Este; Masenge, Andries; Cromhout, M.; Nkukwana, Thobela T.; evm.koster@up.ac.za
There is considerable pressure to eliminate the use of conventional cages in commercial layer hen production systems. However, the assessment of alternative systems that can ensure the hen's ability to perform natural behaviours, while simultaneously enhancing farm productivity and economic efficiency, remains incomplete. This study assessed layer behaviour in a floor system and in enriched cages on a commercial layer farm using human and video-based observations. The study focused on dust bathing, nesting, feather pecking, and perching behaviours, and on the formation of mud balls on the feet. A large proportion (72.9%) of the hens exhibited dust-bathing behaviour, with an average duration of 22.63 minutes. Feather pecking was exhibited by 35.4% of layers in the enriched cages, compared to 58.3% of layers in the floor system. Overall, feather pecking was the least observed behaviour. Layers in enriched cages used perches more (47%) than layers in the floor system (27%), and a negative association was found between body weight and perching in layers in the floor system. At the end of the six-week trial period, 41.67% of the hens had developed mud balls on their toes that exceeded 3 cm in length. The results of this study provide evidence of the relationships between poultry behaviour, welfare, and production. Video-based observations confirmed that farm managers may not be able to identify certain welfare-related behavioural aspects unless they are closely monitored. The results of this study may be used to inform stakeholders about behaviour and welfare considerations in the management of commercial layers.
Identifying the associated risk factors of time to first antenatal care contacts among pregnant women in Somalia : a survival frailty approach
(Bentham Science Publishers, 2025-01) Belay, Denekew Bitew; Fentaw, Kenaw Derebe; Ali, Mahad Ibrahim; Rad, Najmeh Nakhaei; Chen, Ding-Geng (Din)
BACKGROUND : Antenatal care (ANC) utilization offers a wide range of interventions, such as education, counseling, screening, treatment, monitoring, and supporting the health of pregnant women, making it a significant opportunity for expectant mothers. This study aims to investigate the time to the first ANC contact among pregnant women and to identify associated factors by employing the Accelerated Failure Time (AFT) model using different frailty distributions. METHODS : This study used Somalia's Health and demographic survey data. A sample of 3138 women of reproductive age (15-49 years) were included in the study and accelerated failure time (AFT) models with different frailty distributions were compared using information criteria to select the best model. RESULTS : Among the women included in this study, only 33.1% of them received their first ANC contacts within the recommended time during their pregnancy. A gamma frailty model with log-logistic as base-line distribution was found to be the best model for the time-to-first ANC utilization for our data. The final model, based on the log-logistic gamma frailty, identified marital status, mother's occupation, wanted pregnancy, region, parity, wealth index, education level of mother, persons deciding on mother health care, and media exposure are significant (p-value <0.05) predictors of time to the first ANC contact in Somalia. The final model evidenced a high degree of heterogeneity at an individual level regarding the time to the first ANC utilization in Somalia. CONCLUSION : The median time for the first ANC contact among pregnant women was 6.2 months. To ensure accurate analysis and better policy recommendation, different candidate models were compared, and the univariate gamma frailty model with a log-logistic baseline was found to be the most appropriate approach for analyzing time to the first ANC contact among pregnant women. Maternal and child health policies and initiatives must better focus on women's development and implement interventions aimed at increasing the timely initiation of prenatal care services. More specific policy measures, such as targeted educational campaigns, improved pregnancy services, and efforts to minimize regional disparities, should be prioritized as urgent intervention mechanisms.
Meta-analysis of the prevalence of tuberculosis in cattle and zoonotic tuberculosis in humans in sub-Saharan Africa
(BioMed Central, 2025-04) Ngwira, Alfred; Manda, S.O.M. (Samuel); Karimuribo, Esron Daniel; Kimera, Sharadhuli Iddi
BACKGROUND : Tuberculosis (TB) in cattle negatively affects the cattle economy in Africa, with zoonotic TB posing drug-resistance issues in humans. The burden of TB in cattle and zoonotic TB in humans in sub-Saharan Africa (SSA) is not well understood. This study aimed to determine the prevalence of both TB in cattle and zoonotic TB in humans in SSA through meta-analysis. METHODS : Research on TB prevalence was sourced from multiple databases. A random effects meta-analysis model estimated TB prevalence in SSA and its regions, while meta-regression identified risk factors. The analysis included 114 studies for cattle and 59 for humans. RESULTS : The estimated TB prevalence in cattle was 5.06% (95% CI: 3.76–6.78), with a higher burden in West Africa. The prevalence was greater on farms than at abattoirs. Among humans, M. bovis prevalence was 0.73% (95% CI: 0.53–1.01), increasing to 1.56% (95% CI: 1.04–2.33) in TB incident cases, especially in the West and East Africa. Higher prevalence was noted among livestock workers, and in drug-resistant cases. Significant factors influencing TB prevalence varied for cattle and humans, including country, diagnostic methods, and study populations. CONCLUSION : Focusing interventions on farms and livestock workers could help reduce the disease burden.
Factors that affected South African students’ sense of self-perceived academic competence during the COVID-19 lockdown and the rapid transition to online learning
(MDPI, 2025-02) Law-Van Wyk, Eloise; Visser, Maretha; Masenge, Andries; eloise.law-vanwyk@up.ac.za
The COVID-19 pandemic lockdown restrictions had significant impacts on the well-being and academic functioning of students worldwide. When universities closed campuses and moved teaching and learning online, students faced numerous challenges. The researchers conducted a study to establish which factors most affected South African students’ academic competence during the initial months of lockdown and the shift to online learning. Using an online survey that focused on students’ wellness, perceived academic competence and coping behaviour, data were collected from 3239 university students. Multiple linear regression showed that students’ subjective sense of intellectual wellness, coping behaviour, satisfaction with support from the university, and mental health were strong predictors of academic competence. Other factors that influenced students’ academic competence were emotional and spiritual wellness, perceptions of safety and security, and hopefulness. Females, undergraduates, and Faculty of Law students reported higher perceptions of academic competence. These findings have practical implications for universities as they identify factors that contribute to students’ academic competence, especially during times of crisis and online learning. Academic and support services staff at universities may find the findings valuable when developing policies to provide appropriate resources and services to promote and sustain students’ academic functioning.
Prevalence and associated factors of immunization among under-five children in Somalia
(BioMed Central, 2025-03) Belay, Denekew Bitew; Ali, Mahad Ibrahim; Chen, Ding-Geng (Din); Jama, Umalkhair Abdi
BACKGROUND : Children worldwide can live lives free from various illnesses and disabilities due to vaccination. For instance, vaccination has eliminated smallpox, a deformative and frequently fatal illness that claimed an estimated 300 million lives in the twentieth century. However, due to a lack of access to immunization and other health services, 14.3 million infants in 2022 still did not receive their first dose of the Diphtheria-Tetanus-Pertussis (DTP) vaccine, and an additional 6.2 million received only a portion of the scheduled dose. This study aimed to assess prevalence and determinant factors of immunization among under-five children in Somalia using Somalia Health and Demographic Survey (SHDS) Data. METHODS : The study design was cross-sectional, utilizing the SHDS 2020 data. A total of 3916 under-five children who fulfilled the inclusion criteria were included in this study. Count regression models were employed to explore factors associated with the number of vaccinations received per child. RESULTS : In this study, 9.14% of children did not receive any vaccination during their childhood. Different candidate count regression models were compared. Using AIC and BIC, the Negative-binomial (NB) regression model was found to be the best fit. From this model, we found that women ages 20–24 (IRR = 1.192, 95% CI: 1.083, 1.313) and 25–29 (IRR = 1.180, 95% CI: 1.068, 1.305) had a higher number of vaccinations per child compared to women in the 15–19 age group. Women who attended primary education (IRR = 1.090, 95% CI: 1.034, 1.150) and secondary education (IRR = 1.157, 95% CI: 1.058, 1.266) had a higher number of vaccinations per child compared to uneducated women) also correlated with increased vaccination Parity (IRR = 1.090, 95% CI: 1.031–1.153), and wealth quantile (IRR = 1.110, 95% CI: 1.012, 1.217) positively influenced vaccination attendance. Regional disparities were also found to be significant, with Togdheer, Sool, Sanaag, Bari, Nugaal, Bay, Bakool, Mudug, Hiiraan and Galgaduud significantly different from Awdal region. In Negative-Binomial, age, region, residence, educational level, wealth quantile, child size at birth, parity and birth order emerged as key predictors, revealing complex determinants of vaccination utilization in Somalia. CONCLUSIONS : A large proportion of children did not complete the full vaccination schedule. Socio-demographic factors, such as age, region, residence, educational level, wealth quantile, child size at birth, parity, and birth order, had a significant impact on the number of children vaccinated in Somalia. These findings underscore the importance of targeted interventions for addressing these factors. Implementing initiatives based on these conclusions has the potential to enhance vaccination coverage and child health outcomes.
Improved runs-rules precedence charts for monitoring the process location parameter
(Springer, 2025-04) Human, Schalk William; Malela-Majika, Jean-Claude; Kritzinger, Pierre; malela.mjc@up.ac.za
Runs-rules are typically incorporated into control charts to increase their sensitivity to detect small process shifts. However, a drawback of this approach is that the resulting charts are unable to detect large shifts quickly. In this paper, improved runs-rules are introduced to the nonparametric precedence chart to address this limitation by maintaining the sensitivity to small process shifts, while improving the ability to detect large shifts in the process. Performance comparisons between the proposed precedence charts and the precedence charts with standard runs-rules are made in terms of their respective run-length characteristics. The results reveal that the precedence charts with improved runs-rules are superior to the competing charts in detecting large shifts in the process, while maintaining the same sensitivity in the detection of small shifts. A real-life example from the engineering field is given to demonstrate the application and implementation of the new charts.
A power-cardioid candidate for wind direction modelling motivated by two South African case studies
(Springer, 2025-04) Van Wyk-de Ridder, Delene; Rad, Najmeh Nakhaei; Arashi, Mohammad; Ferreira, Johan; Bekker, Andriette, 1958-; johan.ferreira@up.ac.za
Wind energy claims a positive image globally; therefore, accurate modelling of wind direction at generation sites accurately can enhance the potential of this green energy source. The uncertain nature of wind direction can be modelled through probability distributions; in this paper, we propose a flexible yet simple distribution, namely the Power-Cardioid distribution, as an alternative and implementable candidate to model wind direction. After discussing some characteristics, the performance of the Power-Cardioid distribution is evaluated via a simulation study and applied to datasets of two wind farms in South Africa. The numerical results demonstrate that this distribution is a promising and exciting new candidate compared to well-known models within circular statistics.
Development and validation of a knowledge, attitudes and practices (KAP) questionnaire for healthcare professionals on environmental sustainability in healthcare in Southern Africa
(F1000 Research Ltd, 2024-10) Lister, Helga Elke; Mostert, Karien; Botha, Tanita; Field, Emma; Knock, Danté; Mubi, Natasha; Odendaal, Stefani; Rohde, Megan; Maric, Filip
BACKGROUND: The triple planetary crisis of climate change, environmental pollution, and biodiversity loss is increasingly driving poor health outcomes worldwide. Healthcare systems and services are often not environmentally sustainable and compound the problem, while healthcare professionals are also recognised as key leaders in advancing sustainable healthcare. To adopt this leadership position, healthcare professionals’ knowledge, attitudes, and practices regarding environmental sustainability in healthcare must be established. This article reports the development and validation of a new instrument for this purpose that corresponds to the specificities of the Southern African context. METHODS: Questionnaire development followed a seven-stage process. Information was obtained from a 2021 study titled ‘South African Healthcare Professionals’ Knowledge, Attitudes, and Practices Regarding Environmental Sustainability in Healthcare: A Mixed Methods Study’ to develop the instrument. Information was also sourced from the literature regarding environmental sustainability and healthcare to generate the first questionnaire with 29 items. The following stages included two rounds of expert input, separated by a pilot study with the target population to receive feedback regarding the instrument’s structure, relevance, and length. Content validity was determined through statistical analysis. RESULTS: Feedback was received from nine experts in stage two and 13 pilot study participants in stage four and incorporated to improve the questionnaire. In stage six, the questionnaire was rated by seven experts. The content validity index of the questionnaire was calculated at two different stages, after which the indices were compared. Following a final edit, the questionnaire has 24 questionnaire items. The closing analysis calculated the scale content validity index average (S-CVI/Ave) of 0,922; this indicates that the final questionnaire has excellent content validity. CONCLUSION: A questionnaire that assesses the knowledge, attitudes and practices of healthcare professionals regarding environmental sustainability in Southern Africa has been developed and validated. This questionnaire can now be used for further studies in Southern Africa.
Robust PCA with Lw,∗ and L2,1 norms : a novel method for low-quality retinal image enhancement
(MDPI, 2024-07) Likassa, Habte Tadesse; Chen, Ding-Geng (Din); Chen, Kewei; Wang, Yalin; Zhu, Wenhui
Please read abstract in the article.
An optimised rabies vaccination schedule for rural settlements
(Elsevier, 2025-03) Botes, Rian Hendrik; Fabris-Rotelli, Inger Nicolette; Mahloromela, Kabelo; Chen, Ding-Geng (Din); inger.fabris-rotelli@up.ac.za
Please read abstract in the article.
Mortality trends and causes of death in a South African hospital complex pre- and during COVID-19
(AOSIS, 2025-03) Tshabalala, Khanyisile; Fabris-Rotelli, Inger Nicolette; Basu, Debashis; Myburgh, Magriet; Abdullah, Fareed; khanyisile.tshabalala@ up.ac.za
BACKGROUND : Before coronavirus disease 2019 (COVID-19), global health was improving, with declining mortality trends. The pandemic disrupted this progress, increasing mortality in South Africa between April 2020 and March 2022. Pre-pandemic data establishes a baseline for assessing COVID-19’s impact on all-cause mortality. OBJECTIVES : This study examines changes in hospital-based mortality trends in a Gauteng hospital complex from April 2018 to March 2022, addressing the scarcity of such studies during the COVID-19 era. METHOD : A retrospective review of 7815 deaths from April 2018 to March 2022 was conducted. Chi-squared tests were used to analyse deaths by age group and gender, with correlations reported. RESULTS : Mortality rates rose from 3.2% in 2018–2019, peaked at 5.1% in 2020–2021, and declined to 4.2% in 2021–2022. Patients aged 15 years–64 years had the highest death rates, with an increase among those over 65. Male deaths exceeded female deaths, with the smallest difference observed in 2020–2021. Leading causes of death included diseases of the circulatory and respiratory systems, neoplasms, digestive system diseases, and infectious and parasitic diseases. CONCLUSION : The study highlights COVID-19’s impact on mortality, showing variations by year, age, gender, and disease. Contribution: Excess non-COVID-19 deaths likely stemmed from disrupted healthcare services. These findings underscore the need for ongoing monitoring of hospital mortality to identify pandemic-related service disruptions and guide interventions to strengthen healthcare services, improve access to care, and enhance referral systems during unexpected disasters.
Contest competition and injury in adult male sub-Antarctic fur seals
(Springer, 2024-05) Bester, Marthan Nieuwoudt; Rossouw, G.J.; Van Staden, Paul Jacobus; mnbester@zoology.up.ac.za
We recorded intra-sexual behaviour of adult male sub-Antarctic fur seals Arctocephalus tropicalis at Gough Island, Southern Ocean, during the 1975/76 summer breeding season. Our re-analysed data address male ‘contest competition’, which relates to the costs of intra-sexual disputes, including fights. We considered the risks/benefits of fighting through investigation of injuries (n = 353) sustained by adult males (n = 124) in fights. Injuries were predominantly on the forequarters, especially around the insertion areas of the front flippers (41%) with its sparse pelage, compared to the neck and chest areas combined (29%), an area which is well protected by thick pelage. The infliction of serious, sometimes debilitating, injuries to competitors increases a male’s access to females. Injuries predominate in injured, defeated males that gather at non-breeding sites, suggesting that injuries inflicted by dominant males were successful in excluding competing males from breeding sites.
Partial proportional odds of child polio vaccination status among children aged 12–23 Months in Ethiopia
(Wiley, 2025-04) Belay, Alebachew Taye; Belay, Denekew Bitew; Yilema, Seyifemickael Amare; Moyehodie, Yikeber Abebaw
BACKGROUND : The poliovirus is mostly transmitted by the fecal–oral route and can cause viremia after replication in the gastrointestinal tract. The current global coverage of polio immunization is 85%, against the 90% target, while the total coverage of inactivated polio vaccine (IPV) coverage in sub-Saharan Africa is 73%. Only 30% of the children living in rural areas of Ethiopia were fully vaccinated at the appropriate ages in 2016 EDHS (Ethiopian Demographic and Health Survey) data. This study evaluated the prevalence and factors associated with the vaccination status of children aged 12 and 23 months in Ethiopia. METHODS : The Ethiopian Mini Demographic and Health Survey 2019 (Mini EDHS, 2019) was used for this study. The partial proportional ordinal logistic regression model was used to determine the risk factors associated with the polio vaccination status of children aged 12 and 23 months using SAS version 9.40 statistical software at a 5% level of significance. RESULTS : The prevalence of polio vaccination status of children showed that about 33.00% of children never received vaccination, 21.20% were partially vaccinated, and 45.70% were fully vaccinated. The age of mother (AOR [adjusted odds ratio] = 1.06, 95% CI: 1.025, 1.094), age of mother at the first birth (AOR = 0.94, 95% CI: 0.909, 0.976), rural residence (AOR = 0.66; 95% CI: 0.499, 0.867), higher education (AOR = 1.25; 95% CI: 1.241, 1.730), at least four and more ANC (AOR = 2.34; 95% CI: 1.845, 2.968), and ANC visits in health facilities (AOR = 2.33, 95% CI: 1.902, 2.864) had a significant effect on child polio vaccination. CONCLUSION : The coverage of full polio vaccination in Ethiopia was below 50%, which is below the global and regional percentage. Policies and programs must be formulated and implemented with a strong focus on women who have not used ANC visits, residents in rural areas, and uneducated women in child vaccination must be considered in Ethiopia.
The impact of past and current district-level climatic shifts on maize production and the implications for South African farmers
(Springer, 2025-02) Mangani, Robert; Mazarura, Jocelyn; Matlou, Solly; Marquart, Arnim; Archer, Emma Rosa Mary; Creux, Nicky; nicole.creux@fabi.up.ac.za
South Africa’s climate studies generally focus on coarser provincial levels, which aid policy recommendations, but have limited application at the farm level. District level climate studies are essential for farmer participation in climate change mitigation strategies and management. Our study aimed to investigate historical climate data for trends and their influence on maize yields at the magisterial level. Six sites were selected from three major maize-producing provinces in South Africa: Mpumalanga, Northwest, and Free State. Magisterial districts in each province were selected from different Köppen-Geiger climate zones. The climate variables assessed by the Mann–Kendall trend test included maximum or minimum temperature, rainfall, number of extreme high-temperature days, rainfall onset and cessation from 1986 to 2016. The average maximum temperatures were observed to have significant upward trends in most locations, except for Schweizer-Reneke and Bethlehem. The fastest rate of change was observed at Klerksdorp (0.1 °C per 30 years of study), while the Schweizer-Reneke district was the slowest (0.05 °C per 30 years of study). No significant changes were observed in rainfall onset, cessation, or total rainfall in Schweizer-Reneke, Standerton, and Bethlehem, which are scattered across the different provinces. The other districts in each province showed significant changes in these parameters. Rainfall accounted for the significant variation in maize yields over the study period, explaining between 18 and 40% of the variation in the North West, and between 1 and 17% in the Free State. These findings highlight the importance of understanding location-specific changes at a finer scale, which can help farming communities adjust agronomic practices and adapt to local climate shifts.
Gynaecological health patterns and motherhood experiences of female professional football players
(MDPI, 2025-02) Ramagole, Dimakatso Althea; Janse van Rensburg, Dina Christina; Cowie, Charlotte; Mehta, Ritan; Ramkilawon, Gopika Devi; Pluim, Babette M.; Kerkhoffs, Gino M.M.J.; Gouttebarge, Vincent
The aim of this paper is to explore the gynaecological health patterns, contraceptive use, body perception, and motherhood experiences of female professional football players. The participants were recruited via email using FIFPRO (Football Players Worldwide). Online questionnaires were completed by consenting participants. The mean age at menarche was 13.5 years with an average cycle length of 26 days, and a bleeding period of 5 days. Cycle irregularities were experienced by 30%, and menstrual symptoms by 74%. Half of the participants used contraceptives, 60% using hormonal contraceptives, primarily oral contraceptive pills (38%), followed by implants (20%). The body satisfaction score was normal but there was a high drive-for-thinness (DT) score. The motherhood rate was low (1%), with participants experiencing normal conception, vaginal delivery, return to training after 6 weeks, and return to competition after 12 weeks. Our findings are consistent with findings in other elite female athletes with cycle irregularities and a significant number of cycle-related symptoms. The majority of those using contraceptives preferred hormonal contraceptives, reflecting trends seen in other elite athletes. While body satisfaction scores were normal, there was a high DT score, similar to that observed in lean and weight-category sports. The motherhood rate was low, consistent with previous findings in professional football players and other elite athletes. This may be due to a lack of financial support during pregnancy and the post-partum period. FIFPRO and its affiliated unions are negotiating better contracts for female football players.
Identification of common spatial and temporal trends in the epidemiology of cattle bovine tuberculosis and human extrapulmonary and drug-resistant tuberculosis in Malawi
(Elsevier, 2024-12) Ngwira, Alfred; Manda, S.O.M. (Samuel); Karimuribo, Esron D.; Kimera, Sharadhuli Iddi
BACKGROUND: Identification of common spatial disease trends between cattle bovine tuberculosis (BTB) and human extrapulmonary tuberculosis (EPTB) and drug-resistant tuberculosis (DRTB) can support integrated disease control and monitoring programmes. We employed the recently developed multivariate disease mapping methods to examine whether the diseases exhibited any spatial correlation. METHODS: A retrospective study of cattle BTB and human EPTB and DRTB cases from 2018 to 2022 was conducted. Bivariate shared spatiotemporal components models were fitted to a) cattle BTB and human EPTB and b) cattle BTB and human DRTB at the district level in Malawi, with cattle density, human density and climatic variables as independent variables. RESULTS: Disease specific spatial effects were higher in the southern half of the country, while the shared spatial effects were more dominant in both the south and western parts of the country. The shared temporal effects showed constant trends, while disease specific temporal effects showed an increasing pattern for cattle BTB and a constant pattern for human EPTB and DRTB. The predicted disease incidence pattern for all forms of TB in the period without data showed a constant pattern over the years. Cattle density was positively associated with cattle BTB (: 0.022; 95% Credible Interval (CI): 0.004, 0.042). Human density was positively associated with human EPTB (: 0.005; 95% CI: 0.001, 0.009). CONCLUSION: Cattle BTB and human EPTB and DRTB have a common spatial pattern in the west and southern parts of Malawi. Integrated interventions targeting high-density areas for cattle and human may have positive impacts on cattle BTB and human EPTB and DRTB.
Alternative skew Laplace scale mixtures for modeling data exhibiting high-peaked and heavy-tailed traits
(Springer, 2024-11) Otto, A.F.; Bekker, Andriette, 1958-; Ferreira, Johannes Theodorus; Arslan, O.; johan.ferreira@up.ac.za
The search and construction of appropriate and flexible models for describing and modelling empirical data sets incongruent with normality retains a sustained interest. This paper focuses on proposing flexible skew Laplace scale mixture distributions to model these types of data sets. Each member of the collection of distributions is obtained by dividing the scale parameter of a conditional skew Laplace distribution by a purposefully chosen mixing random variable. Highly-peaked, heavy-tailed skew models with relevance and impact in different fields are obtained and investigated, and elegant sampling schemes to simulate from this collection of developed models are proposed. Finite mixtures consisting of the members of the skew Laplace scale mixture models are illustrated, further extending the flexibility of the distributions by being able to account for multimodality. The maximum likelihood estimates of the parameters for all the members of the developed models are described via a developed EM algorithm. Real-data examples highlight select models’ performance and emphasize their viability compared to other commonly considered candidates, and various goodness-of-fit measures are used to endorse the performance of the proposed models as reasonable and viable candidates for the practitioner. Finally, an outline is discussed for future work in the multivariate realm for these models.
Hospital accessibility catchment areas as a fuzzy lattice data structure
(Copernicus Publications, 2024-10) De Klerk, Michelle; Fabris-Rotelli, Inger Nicolette; inger.fabris-rotelli@up.ac.za
The accessibility to basic facilities and services plays a pivotal role in every society and city planning. Spatial accessibility can vary between cities and countries and is mainly defined by the ease at which facilities can be accessed by communities. Facilities can provide essential services and/or products such as pharmacies, clinics, schools, universities, etc. Spatial accessibility is dependent on the spatial impedance between a facility and the target population and can be illustrated with catchment areas. We propose a fuzzy lattice catchment area method which uses a semi-supervised learning algorithm to create overlapping catchment areas. This methodology is applied to determine the accessibility to hospitals in South Africa and provides an illustration on the difference for regions with high accessibility compared to low accessibility. The application can easily be adapted in a variety of fields based on industry type, drive-time thresholds, supply capacity and the target population.
Exploring the effect of road network structure on inter-regional accessibility in a diverse road network
(Copernicus Publications, 2024-10) Thiede, Renate Nicole; Fabris-Rotelli, Inger Nicolette; Debba, Pravesh; Cleghorn, Christopher; inger.fabris-rotelli@up.ac.za
Accessibility analyses quantify the level of access to certain areas or opportunities, such as employment and healthcare facilities. Since public data is often aggregated at the level of regions, such as administrative units, it is useful to quantify accessibility between regions. Many factors influence inter-regional accessibility, most notably the accessibility metric used, and the way in which regions are chosen. This paper investigates the effects of road network structure on accessibility, using a previously developed inter-regional accessibility model that bases its accessibility metric on travel distance via the road network. This paper considers an area within the City of Tshwane municipality in South Africa. We investigate the effects of road structure in two ways. Firstly, regions are chosen based on the road network structure, which is done by extending a previously developed road network clustering algorithm for this novel use. Different spatial scales of regionalisation are considered, and the accessibility between these regions is compared to the accessibility between administrative units within the study area. Secondly, the effect of road network homogeneity on accessibility is investigated, where homogeneity corresponds to a uniform concentration of roads across a region. The results show that although road network homogeneity does not significantly correlate with accessibility, the way in which regions are chosen and their spatial scale has a strong effect on the results of the accessibility model. Our novel method of obtaining regions thus provides fresh insights into road-based accessibility within the City of Tshwane.
Assessing classification performance for sampled remote sensing data
(Copernicus Publications, 2024-10) Rangongo, Tshepiso Selaelo; Fabris-Rotelli, Inger Nicolette; Thiede, Renate Nicole; renate.thiede@up.ac.za
Big data poses challenges for storage, management, processing, analysis and visualisation. One technique of handling big data is the use of a representative sample of the data. This paper proposes a sampling algorithm which makes use of multivariate stratification with the aim of obtaining a sample that best represents the population while minimising the number of images in the sample. The proposed sampling algorithm performs effectively on a big spatial image dataset of crop types. The results are assessed by measuring the number of images sampled and as well as matching the proportionality of the population crop percentages. The samples obtained from the proposed algorithm are then used for land cover classification. An ensemble method called random forest is trained on the samples and accuracy is assessed. Precision, recall and F1-scores per crop type are computed as well as the overall accuracy. The random forest classifier performed best on the proposed sample with the least number of images. In addition, the classifier performed better on the proposed sample than it did on a random sample as the proposed sample due to the more informative data. This research develops an effective way of sampling big data for crop classification.

Browse

Recent Submissions