Reproduction code for "Who should fold the laundry? A Bayesian quality scheduling index for evaluating the household adoption of service robots"

Python code that exactly reproduces every quantitative result in the associated manuscript (submitted to Socio-Economic Planning Sciences): the household-level net present value Monte Carlo across three robot price tiers (Table 1, Section 3.2), the Bayesian learning experiment comparing a Thompson-sampling scheduler against a static allocator over a 90-day horizon (Section 3.3), and the three-part sensitivity analysis — Spearman rank correlations, prior-perturbation scenarios, and a one-at-a-time tornado analysis (Table 2, Fig. 1, Section 3.5). Two identical implementations are provided: bqsi_sim.py (plain Python script) and bqsi_sim.ipynb (Jupyter/Google Colab notebook). All random draws use fixed seed 42, so outputs are deterministic. No empirical data are used or included; all results are simulation outputs under the documented priors. Requirements: Python 3 with NumPy; matplotlib is needed only for the optional figure cell.