In class Gallery Setup Planning Project

In-Class Subgroup Sprint (50 minutes)

Goal: In groups of 3, brainstorm one key part of the gallery-planning algorithm problem for 25 minutes, capturing your work in a shared Google Sheet. Then we’ll do 6-minute reports (3 groups = ~18 minutes) plus discussion.

Schedule

  • 0–2 min: Form groups of 3, pick a focus area (A/B/C below).
  • 2–5 min: Create your Google Sheet and share it with your group + instructor.
  • 5–30 min: Brainstorm + fill out the sheet.
  • 30–48 min: Group reports (6 min each). Stick to your sheet.
  • 48–50 min: Wrap-up: we combine outputs into one class “spec”.

Google Sheet Instructions (everyone)

  1. Create a new Google Sheet named: Gallery Planning – Group [A/B/C] – [Your Names]
  2. Share with:
    • All group members (edit access)
    • Instructor (edit access)
  3. Create tabs (sheets) exactly as listed in your focus area below.
  4. During your report, screen share your sheet and use it as your outline.

Group A Focus: Data + Constraints Model

Prompt: What are the inputs to the problem, what constraints must always hold, and what outputs should an algorithm generate?

Make these tabs in your Google Sheet

Tab 1 — Inputs

Create a table with these columns (add rows with concrete examples):

  • Item (Artwork / Room / Wall / Pedestal / Lighting zone / Path segment)
  • ID/Name
  • Attributes (size, medium, weight, fragility, theme tags, required clearance, etc.)
  • What we can decide (placement, orientation, grouping, order, label position)
  • Notes

Tab 2 — Hard Constraints

Table columns:

  • Constraint
  • What it applies to
  • How to check it (algorithmically)
  • Example violation

Examples to consider: wall length limits, min spacing between works, sculpture stability, max weight per pedestal, accessibility clearances, don’t block exits, keep walkways passable, lighting requirements, security restrictions.

Tab 3 — Soft Constraints / Preferences

Same format as hard constraints, but note how it becomes a “penalty” rather than a strict rule.

Tab 4 — Output Spec

Define what a “layout solution” looks like in data form. Table columns:

  • Artwork ID
  • Assigned location (room + wall segment / pedestal)
  • Orientation (if relevant)
  • Group/Theme label
  • Notes

Report (6 minutes)

  • Your proposed minimum viable input dataset
  • Top 5 hard constraints and how to check them
  • What your solution output format should be

Group B Focus: Objective Function + Evaluation

Prompt: If two layouts are both valid, how do we decide which is better? Define measurable goals and a scoring function.

Make these tabs in your Google Sheet

Tab 1 — Goals & Metrics

Table columns:

  • Goal (e.g., thematic coherence, balance, flow)
  • Metric (how you’d measure it)
  • Scale (0–1, 0–100, counts, etc.)
  • Data needed
  • Tradeoffs (what it might conflict with)

Tab 2 — Scoring Function Draft

Propose a scoring formula with weights. Table columns:

  • Component
  • Definition
  • Weight
  • Penalty/Bonus examples

Tip: Include hard constraint penalties as “infinite”/disqualifying, then sum weighted soft metrics.

Tab 3 — Test Cases

Design 3–5 “toy” scenarios to evaluate algorithms. Table columns:

  • Scenario
  • What’s challenging
  • What a good solution should look like
  • What metrics should reveal

Report (6 minutes)

  • Your best 6–10 metrics (and what data they require)
  • A concrete scoring function proposal (with weights)
  • One test case you think will expose bad heuristics

Group C Focus: Algorithm Strategy + Heuristics

Prompt: The search space is huge. Propose a practical algorithm approach and the heuristics/moves that make it work.

Make these tabs in your Google Sheet

Tab 1 — Search Space + Representation

Table columns:

  • Decision variable (what we choose)
  • Possible values
  • Why it explodes combinatorially
  • Representation (how we store a state)

Tab 2 — Candidate Algorithms

List 2–3 approaches and when you’d use them. Table columns:

  • Approach (greedy, backtracking, local search, annealing, etc.)
  • State / partial solution
  • How it makes choices
  • Pros / Cons

Tab 3 — Heuristics & Move Operators

Table columns:

  • Heuristic / Move
  • What it does
  • Why it helps
  • How to compute it fast

Examples: place largest pieces first, avoid clustering heavy themes, keep high-interest works near entrances, penalize tight bottlenecks, swap two artworks, shift a block along a wall, rotate a painting, reassign one theme cluster to another room.

Tab 4 — Pruning / Speedups

Table columns:

  • Pruning rule / speedup
  • When it triggers
  • Why it’s safe (or acceptable)

Report (6 minutes)

  • Your recommended algorithm approach (and why)
  • Your top heuristics + move operators
  • One concrete idea to make it efficient

Quality bar

Your sheet should be readable by another group that did a different focus area. Use short phrases and concrete examples. If you’re stuck, add a row called “Assumption” and write what you’re assuming.

Scroll to Top