Deterministic Preprocessing and Interpretable Fuzzy Banding for Cost-per-Student Reporting

A recent technical document on arXiv (2603.04905v2) introduces a systematic, rule-based approach for converting administrative spreadsheet exports into cost-per-student analyses. This process is executed through a Python script named cad_processor.py, which processes a Casual Academic Database (CAD) export workbook, consolidating inclusive on-costs and student numbers into totals for subject-year and school-year. It calculates cost-per-student ratios from these aggregates. The output includes a workbook with four tabs: Processing Summary (which logs runs and counters), Trend Analysis (matrix of school-year cost-per-student), Report (comprehensive subject-level table), and Fuzzy Bands (annual anchors, membership weights, and labels). The run record features a SHA-256 hash for input workbook traceability. The paper highlights the necessity of deterministic preprocessing for clarity in budgeting, workload assessments, and governance discussions, where exported workbooks are often utilized as reference points. Additionally, fuzzy banding offers understandable categories for cost-per-student data, supporting informed decision-making. This paper falls under the computer science and information retrieval categories on arXiv.