Kazi Md. Anzum

Mr. Kazi Md. Anzum

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 January 2026 in Global Journal of Flexible Systems Management
Verified
Journal Article
Closed access
PDF file
TCTalat Mahmud Chowdhury
Talat Mahmud Chowdhury
KAKazi Md. Anzum
Kazi Md. Anzum

29 March 2026 in Circular Economy and Sustainability
Verified
Journal Article
Closed access
SSSamiran Sardar
Samiran Sardar
PAPartho Das Antor
Partho Das Antor
KAKazi Md. Anzum
Kazi Md. Anzum
ASAnn Nazmus Shakib
Ann Nazmus Shakib
KSKazi Md. Shahriar
Kazi Md. Shahriar

1 January 2026 in Journal of Engineering
Verified
Journal Article
Open access
KAKazi Md. Anzum
Kazi Md. Anzum
MIMd. Saiful Islam
Md. Saiful Islam
MPMahmud Parvez
Mahmud Parvez
Md. Mahbubur RahmanMd. Mahbubur Rahman
Md. Mahbubur Rahman
Outpatient clinics frequently experience delays because walk‐in arrivals and patient no‐shows occur unpredictably, disrupting patient flow and limiting the efficient use of clinical resources. This study improves appointment scheduling under such uncertainty by reducing the time patients spend in the clinic and increasing the number of patients served. We developed the walk‐in and no‐show patient appointment scheduling system (WINS‐PASS), which integrates particle swarm optimization (PSO) with discrete‐event simulation (DES). The hybrid framework evaluates candidate appointment templates under random walk‐ins, no‐show behavior, and variable service times, and it generates a strong Pareto frontier that balances two objectives: minimizing total service time (TST) and maximizing the number of patients served (NOP). Across booking volumes, the PSO‐based schedules reduce the aggregate TST by approximately 17%–26% and increase daily throughput by approximately 7%–13%, without overtime. The framework also outperforms three established multiobjective algorithms: SPEA2 attains 170 min of TST with 31 patients, NSGA‐II attains 175 min with 29 patients, and IBEA attains 180 min with 22 patients, whereas PSO achieves the best‐balanced outcome of 151 min with 36 patients. Pairwise Wilcoxon rank‐sum tests over 100 replications confirm that these differences are statistically significant. Sensitivity analysis shows that morning walk‐ins improve patient flow, moderate no‐show rates support stable operation, and the mean service time strongly influences efficiency. These findings help clinics design reliable daily schedules that reduce delays and improve patient access while using existing resources more effectively, without additional staffing or overtime.

23 August 2024 in Heliyon
Verified
Journal Article
Open access
KAKazi Md. Anzum
Kazi Md. Anzum
GKGolam Kibria
Golam Kibria
In the realm of healthcare, an imperative necessity for all, institutions are increasingly recognizing the advantages of adopting lean strategies to enhance performance. Lean implementation in healthcare can lead to significant improvements in efficiency, patient care, and overall institutional performance. This paper aims to assess the readiness levels for implementing lean practices in healthcare institutes in Bangladesh, employing a fuzzy logic approach. The construction of a conceptual model is grounded in literature review and expert opinions, incorporating critical enablers, criteria, and attributes identified from extensive research. Factors measured include leadership commitment, workforce capability, operational processes, technological infrastructure, and organizational culture, each pivotal in determining readiness for lean implementation. The fuzzy logic approach is particularly useful in this context due to its ability to handle uncertainty and imprecision, which are common in complex environments like healthcare. This methodology not only provides a clear picture of current capabilities but also highlights specific areas that need enhancement, paving the way for more targeted and effective lean interventions. Data sourced from consultations with experts in three prominent hospitals in Bangladesh forms the basis of the analysis, enabling a detailed examination of readiness levels. The model's application of fuzzy logic facilitates a comprehensive assessment, revealing 12 critical attributes across the hospitals that require attention. Interestingly, the evaluation identifies varying levels of readiness, with two hospitals demonstrating moderate readiness and one showing a lower level. This conceptual approach has significant potential to assist top management in healthcare companies by providing a structured framework to prioritize crucial areas for improvement. By accurately assessing readiness levels and pinpointing weaker aspects before implementing lean strategies, this study aims to transform the healthcare industry. Ultimately, its implementation has the potential to enhance organizational performance and elevate standards in patient care, contributing to improved healthcare delivery in Bangladesh and beyond.

21 December 2024
Verified
Conference Paper
Open access
PDF file
NMNasif Morshed
Nasif Morshed
MRMushfiqur Rahman
Mushfiqur Rahman
KAKazi Md. Anzum
Kazi Md. Anzum
This study proposes a novel framework to manage organizational agility in uncertain environments while exploring the interplay between organizational culture, disruption severity, and AI readiness and responsiveness to emerging types of disruption. The research draws its insights from the perspective of dynamic capabilities and asserts that organizational culture has the ability to guide AI integration at junctions when disruptions occur. It was based on a survey methodology of 383 respondents across 4 sectors – manufacturing, retail, IT services and healthcare. To validate hypothesized relationships, Partial Least Squares Structural Equation Modeling (PLS-SEM) was used. The results confirmed that organizational culture has a significant positive impact on AI readiness and responsiveness. Additionally, the results indicate that disruption severity is a key moderator, enhancing the effect of organizational culture in these particularly disrupted contexts. This shows two roles of organizational culture in ensuring technological capability as proactive and reactive. In this study, organizational culture and disruption management was unified within a framework to deliver unique insights into the strategic alignment of cultural and technological resources. To extend existing literature, and address gaps in the existing literature, the research considers the moderating role of disruption severity in dynamic environments. This is actionable guidance for organizations to better develop their technological adaptability, especially in a crisis. Future research on cultural dimensions as well as longitudinal impacts could enrich understanding as much as possible. The novel contributions of the study make it valuable to both academic and practitioner audiences interested in understanding organizational behavior and technology management.

Verified
Open access
PDF file
JMJarin Tasnim Mumu
Jarin Tasnim Mumu
IHIsmail Hossain
Ismail Hossain
KAKazi Md. Anzum
Kazi Md. Anzum
Chronic kidney disease (CKD) is a major global health problem, and early detection is important for reducing complications and improving treatment. Machine learning has shown strong potential for CKD prediction, but many existing studies rely on single datasets, face severe class imbalance, and provide limited interpretability. These challenges reduce model reliability and make cross-dataset use in real clinical settings more difficult. Therefore, this study aimed to develop an interpretable and adaptable machine learning framework for CKD risk prediction across different datasets. The proposed hybrid explainable machine learning (HXAI-ML) framework used two public CKD datasets: the UCI CKD Risk Factor Prediction dataset as the source dataset and the Kaggle CKD dataset as the target dataset. After preprocessing, label binarization, imputation, and standard scaling, seven balancing methods and six classifiers were evaluated. A target adaptation setting was applied by using 10% of the target dataset for calibration and 90% for blind testing. SHAP, LIME, and Permutation Importance Analysis were used to explain the model predictions. Among all tested combinations, RO + TL + ETC achieved the highest overall accuracy among the evaluated target-domain adaptation configurations and was therefore selected as the primary model under the accuracy-based selection criterion. On the 1,494-sample blind holdout it obtained 91.97% accuracy, 92.24% precision, 99.64% recall, 95.80% F1-score, and 67.52% ROC-AUC. However, its specificity of 5.74% and balanced accuracy of 52.69% indicate substantial class-wise imbalance in predictive performance. The findings demonstrate the methodological feasibility of cross-dataset target-domain adaptation and explainable machine learning under limited target-domain data. However, the relatively low ROC-AUC values and low specificity for the non-CKD class indicate limited discriminative ability. Therefore, the proposed framework should be regarded as an exploratory cross-dataset evaluation rather than a clinically deployable screening or decision-support system.