Claim Missing Document
Check
Articles

Found 2 Documents
Search

Optimization Principles in Pedagogical Method Selection: A Topology Optimization-Inspired Framework for Teaching STEM Concepts in Indonesian High Schools Muhammad Ihsan; Andri Afrizal; Andy Prasetyo Wati
JURNAL HURRIAH: Jurnal Evaluasi Pendidikan dan Penelitian Vol. 6 No. 4 (2025): Jurnal Hurriah: Journal of Educational Evaluation and Research
Publisher : Yayasan Pendidikan dan Kemanusiaan Hurriah Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56806/jh.v6i4.400

Abstract

Pedagogical optimization frameworks are becoming more popular as developed countries struggle to improve STEM (Science, Technology, Engineering and Mathematics) education quality. This systematic review combines STEM teaching method selection and optimization approachment research. We investigate theoretical foundations, methodological approaches, empirical data, and implementation issues from 2010–2024 peer-reviewed articles. Cognitive learning theories (including Bloom's taxonomy and constructivism), effectiveness research on teaching methods, operations research optimization techniques, and resource-limited educational constraints are all examined in our analysis. The findings show numerous results. First, active learning methods outperform traditional instruction (effect sizes from d=0.40 to d=0.75), but their efficacy varies throughout Bloom's taxonomy levels. Optimization frameworks are useful for timetabling and resource allocation, however pedagogical technique selection is underdeveloped. Third, resource limits in developing countries require context-adapted solutions, not just scaled-down versions of well-resourced ways. Fourth, engineering optimization methods in educational science are promising but have gotten little attention. A review finds several serious shortcomings. Optimization methodologies are rarely empirically validated in teaching situations. Research on simultaneous optimization across several learning objectives with realistic constraints is uncommon. These deficiencies include evidence-based optimization frameworks, rigorous testing across varied settings, and substantial scalability and cost-effectiveness research, especially in resource-limited contexts.
A Systematic Review of IoT-Enabled Predictive Drying for Arabica Coffee: Toward a Python-Based Digital Twin Framework for the Gayo Highlands, Aceh Tengah, Indonesia Muhammad Ihsan; H. Susanto; Nuzuli Fitriadi
Jurnal Inotera Vol. 11 No. 2 (2026): July - December 2026
Publisher : LPPM Politeknik Aceh Selatan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31572/inotera.Vol11.Iss2.2026.ID721

Abstract

Post-harvest drying is the single most decisive determinant of Arabica coffee quality, yet in the Gayo Highlands of Aceh Tengah (950 - 1,650 m.a.s.l.) it remains an experience based, weather dependent operation conducted largely on tarpaulins and open patios. High relative humidity, frequent rainfall, and low ambient temperature routinely prolong drying beyond ten days, produce non uniform final moisture content, and expose parchment to fungal colonisation and ochratoxin risk, causing smallholders to fail the 12.5% maximum moisture threshold of SNI 01-2907-2008 and to forfeit specialty price premiums. This systematic literature review, conducted following PRISMA guidelines on peer reviewed publications from 2019–2026, synthesises three research streams that have so far evolved in isolation: (i) thin layer drying kinetics of parchment coffee, (ii) Internet of Things (IoT) architectures for solar and hybrid dryers, and (iii) digital twin methodology in food and grain drying. The review finds that existing coffee drying IoT deployments are overwhelmingly reactive while validated kinetic models (modified Midilli, Page) capable of such prediction remain confined to laboratory dryers and are never coupled to field sensor streams. Digital twin research in drying has concentrated on grain and fruit, leaving coffee, and highland smallholder contexts in particular, unaddressed. From this synthesis the review derives seven research gaps and proposes SIKOPI-DT, a contextualised five layer Python based digital twin framework in which a modified Midilli kinetic core is continuously reparameterised by low cost IoT sensors (ESP32, SHT31, load cell) to forecast the time to target moisture content and to drive predictive supplementary-heat control under Gayo weather scenarios. Synthesised evidence indicates that such a system can plausibly reduce drying time by 30–50%, cut specific energy consumption by 15–30%, and improve moisture uniformity relative to open-sun practice. A three phase experimental validation roadmap and a smallholder economic feasibility analysis are presented, positioning this review as the theoretical foundation for a physical prototype in Aceh Tengah.