Energy use in public buildings
Personal proposal by Per-Arne Andersen
D2 · Spring 2027
https://thesis.uya.no/proposals/a-smarter-building-day/

Forecast building electricity use and show facility managers where forecasts are unreliable.

TECHNICAL
- Select buildings using 2016 coverage only; flag missing readings.
- Fit ridge regression using hour/day and past readings; build an actual-versus-forecast chart.

DATA (public)
Building Data Genome 2: electricity, 2016–2017
First 10 office buildings by ID with ≥90% observed 2016 readings. Publish IDs before opening 2017 outcomes. Train on 2016; score observed 2017 hours.
Building Data Genome 2: https://github.com/buds-lab/building-data-genome-project-2

METHOD
Train ridge regression on 2016; compare day-ahead forecasts with last-week values on observed 2017 targets. Report per-building MAE and coverage. Six facilities staff review 6 disjoint reports with/without missing-data labels; score gap recognition and verification decisions.

OUTPUT
Forecast dashboard, data-preparation script and benchmark.

BACHELOR
Task: Build a building-energy dashboard with the established same-hour-last-week forecast.
Data: Building Data Genome 2: electricity, 2016–2017 (public)
First 10 office buildings by ID with ≥90% observed 2016 readings. Publish IDs before opening 2017 outcomes. Train on 2016; score observed 2017 hours.
Requires: Recruit facility managers for the decision study; no energy-saving claim.
Method: Use the specified BDG2 meters; report forecast error and missing-data coverage, then observe facilities tasks. Plan 4 participant sessions.
Output: A reproducible dashboard with a basic forecast evaluation.
Use relevant literature to justify the established approach; a new research contribution is not the aim of this proposal.

MASTER
Investigate when a better energy forecast produces a better facilities decision. Compare the specified forecasting baselines and isolate missing-data disclosure in the interface; analyse forecast error separately from justified action.
Intended contribution: Evidence connecting data quality, forecast evaluation and operational interpretation.

STARTING PAPERS
Wang & Strong (1996) — Beyond Accuracy: https://www.tandfonline.com/doi/abs/10.1080/07421222.1996.11518099
Distinguish accuracy, completeness, representation and fitness for the task.
Vessey & Galletta (1991) — Cognitive Fit: https://pubsonline.informs.org/doi/10.1287/isre.2.1.63
Test whether a representation helps one kind of task more than another.

Search Scopus or Web of Science and ACM Digital Library using the topic query, then follow citations to the thesis start date. Record searches and compare methods, data, findings and limitations in literature-matrix.csv. Use that review to confirm or revise the gap and choose a current comparator. The linked papers are starting points.

START HERE
Tools: Python, pandas, scikit-learn, Streamlit
1. Download BDG2 metadata and electricity readings; record the repository revision.
2. Select office buildings by 2016 coverage only and freeze their IDs before examining 2017 outcomes.
3. Plot one building and evaluate last-week forecasting before fitting regression.
Literature search: building energy forecast facility manager information quality
Study controls:
- Use raw meter data; select buildings using 2016 only. Check that 10 qualifying office buildings exist before fixing scope.
- Features: hour, weekday, lags 24 and 168 hours. Tune on chronological 2016 folds only.
- Use only readings available at forecast issue time. Share a training-derived missing-lag fallback across methods.
- Keep forecast values identical in the labelled/unlabelled report comparison.
- Before collecting participant data, agree consent, storage and withdrawal handling with the supervisor. Use participant codes, not names, in study files.
- For the master’s study, use the research task above to define the factors and comparisons in this pilot plan. Preregister one primary outcome and feasible scope after the literature review; do not add every possible model or interface variant.

REQUIRES
Recruit 6 facility managers for the decision study; no energy-saving claim.

--------------------

NON-TECHNICAL
- Prepare monthly energy reports for three buildings.
- Ask facility managers to identify unusual consumption and explain what they would investigate.

DATA (mixed)
BDG2 reports + 6 facility-manager interviews
First 3 office buildings selected using ≥90% 2016 coverage only. Prepare 6 static 2017 report pairs with spreadsheet-produced weekly forecasts; missing-data labels differ.
Building Data Genome 2: https://github.com/buds-lab/building-data-genome-project-2

METHOD
Balanced-order report review; record issues noticed and proposed actions. Code how missing data changes confidence in the report.

OUTPUT
Revised report template and findings from the 6 interviews.

BACHELOR
Task: Find how facilities staff interpret energy charts and missing readings.
Data: BDG2 reports (mixed)
First 3 office buildings selected using ≥90% 2016 coverage only. Prepare 6 static 2017 report pairs with spreadsheet-produced weekly forecasts; missing-data labels differ.
Requires: Recruit facility managers; reports must be prepared.
Method: Use static BDG2 charts and spreadsheet forecasts in 4 walkthroughs; record misunderstood labels and decisions.
Output: A revised report layout and practical data-quality guidance.
Use relevant literature to justify the established approach; a new research contribution is not the aim of this proposal.

MASTER
Explain when staff disregard an energy warning despite accurate numbers. Compare quality judgements and task fit across missing-data/chart scenarios; check whether role obligations better explain apparent distrust.
Intended contribution: A theory-based account of how energy reports become actionable.

STARTING PAPERS
Wang & Strong (1996) — Beyond Accuracy: https://www.tandfonline.com/doi/abs/10.1080/07421222.1996.11518099
Distinguish accuracy, completeness, representation and fitness for the task.
Orlikowski & Gash (1994) — Technological Frames: https://dl.acm.org/doi/10.1145/196734.196745
Compare how roles interpret the purpose, operation and use of the same system.

Search Scopus or Web of Science and ACM Digital Library using the topic query, then follow citations to the thesis start date. Record searches and compare methods, data, findings and limitations in literature-matrix.csv. Use that review to confirm or revise the gap and choose a current comparator. The linked papers are starting points.

START HERE
Tools: LibreOffice Writer/Calc, audio recorder with consent; no programming required.
1. Prepare a pilot with 2 examples from: BDG2 reports + 6 facility-manager interviews. Write the task questions and a reference answer sheet.
2. Write a recruitment message, information sheet and consent form for the participants named above. Agree privacy handling with the supervisor before contact.
3. Pilot one session after approval; revise unclear questions, freeze the task sets and coding categories, then recruit the planned sample.
Literature search: building energy forecast facility manager information quality qualitative scenario study
Study controls:
- Hold predictions and report contents constant; each participant sees one version of each report.
- Compare confidence, reasons and intended action; do not infer actual energy savings.
- Before collecting participant data, agree consent, storage and withdrawal handling with the supervisor. Use participant codes, not names, in study files.
- Pilot separately, then freeze the questions and coding plan. Check objective answer keys independently; keep an audit trail of coding, including disagreements.
- For comparisons, counterbalance order and case assignment; do not show a person both versions of one case. Report participant-level findings, not repeated tasks as independent people.

REQUIRES
Recruit facility managers; reports must be prepared.

Bachelor: apply established methods and evaluate a practical solution or study. Master: position a research question in current scientific literature, investigate a mechanism or unresolved problem, and explain the contribution. Final scope is agreed with me.
Study sizes are proposed; participant recruitment and planned materials are not already arranged.