add dual plot for weekday or weekends

src/patientflow/viz/arrival_rates.py

@@ -434,3 +434,191 @@ def plot_cumulative_arrival_rates(
         filename = f"{file_prefix}{clean_title_for_filename(title)}"
         plt.savefig(media_file_path / filename, dpi=300)
     plt.show()
+
+
+def plot_dual_arrival_rates(
+    inpatient_arrivals_1,
+    inpatient_arrivals_2,
+    title,
+    labels=("Set 1", "Set 2"),
+    lagged_by=None,
+    curve_params=None,
+    time_interval=60,
+    start_plot_index=0,
+    x_margin=0.5,
+    file_prefix="",
+    media_file_path=None,
+):
+    """
+    Plot two sets of inpatient arrivals on the same chart for comparison.
+    When spread rates are present, they use the main colors while other rates are grey.
+    """
+    # Calculate arrival rates for both sets
+    def process_dataset(arrivals):
+        rates_dict = time_varying_arrival_rates(arrivals, time_interval)
+        rates, hour_labels, hour_values = process_arrival_rates(rates_dict)
+
+        if lagged_by is not None:
+            lagged_dict = time_varying_arrival_rates_lagged(arrivals, lagged_by, time_interval)
+            rates_lagged, _, _ = process_arrival_rates(lagged_dict)
+        else:
+            rates_lagged = None
+
+        if curve_params is not None:
+            x1, y1, x2, y2 = curve_params
+            spread_dict = true_demand_by_hour(arrivals, x1, y1, x2, y2)
+            rates_spread, _, _ = process_arrival_rates(spread_dict)
+        else:
+            rates_spread = None
+
+        return rates, rates_lagged, rates_spread, hour_labels, hour_values
+
+    # Process both datasets
+    rates1, rates1_lagged, rates1_spread, hour_labels, hour_values = process_dataset(inpatient_arrivals_1)
+    rates2, rates2_lagged, rates2_spread, _, _ = process_dataset(inpatient_arrivals_2)
+
+    # Helper function for cyclic data
+    def get_cyclic_data(data):
+        return data[start_plot_index:] + data[0:start_plot_index]
+
+    # Create plot
+    plt.figure(figsize=(12, 7))
+    x_values = get_cyclic_data(hour_labels)
+
+    # Define main colors for each group
+    color1 = '#1f77b4'  # Blue
+    color2 = '#9467bd'  # Purple
+
+    # Define grey shades
+    grey1 = '#888888'  # Darker grey
+    grey2 = '#BBBBBB'  # Lighter grey
+
+    # Plot first dataset
+    y_values1 = get_cyclic_data(rates1)
+    if rates1_spread is not None and len(rates1_spread) > 0:
+        # Base arrival rates in grey
+        plt.plot(
+            x_values,
+            y_values1,
+            marker="x",
+            color=grey1,
+            markersize=4,
+            linestyle=":",
+            linewidth=1,
+            label=f"{labels[0]} - Arrivals",
+        )
+
+        # Lagged arrival rates in lighter grey
+        if rates1_lagged is not None:
+            plt.plot(
+                x_values,
+                get_cyclic_data(rates1_lagged),
+                marker="o",
+                markersize=4,
+                color=grey2,
+                linestyle="--",
+                linewidth=1,
+                label=f"{labels[0]} - Lagged arrivals",
+            )
+
+        # Spread arrival rates in main color
+        plt.plot(
+            x_values,
+            get_cyclic_data(rates1_spread),
+            marker="o",
+            color=color1,
+            label=f"{labels[0]} - Aspirational ({int(curve_params[1]*100)}% in {int(curve_params[0])}h)",
+        )
+    else:
+        # If no spread rates, use normal coloring
+        plt.plot(
+            x_values,
+            y_values1,
+            marker="o",
+            color=color1,
+            linestyle="-",
+            label=f"{labels[0]} - Arrivals",
+        )
+
+        if rates1_lagged is not None:
+            plt.plot(
+                x_values,
+                get_cyclic_data(rates1_lagged),
+                marker="s",
+                color=color1,
+                linestyle="--",
+                label=f"{labels[0]} - Lagged arrivals",
+            )
+
+    # Plot second dataset
+    y_values2 = get_cyclic_data(rates2)
+    if rates2_spread is not None and len(rates2_spread) > 0:
+        # Base arrival rates in grey
+        plt.plot(
+            x_values,
+            y_values2,
+            marker="x",
+            color=grey1,
+            markersize=4,
+            linestyle=":",
+            linewidth=1,
+            label=f"{labels[1]} - Arrivals",
+        )
+
+        # Lagged arrival rates in lighter grey
+        if rates2_lagged is not None:
+            plt.plot(
+                x_values,
+                get_cyclic_data(rates2_lagged),
+                marker="o",
+                markersize=4,
+                color=grey2,
+                linestyle="--",
+                linewidth=1,
+                label=f"{labels[1]} - Lagged arrivals",
+            )
+
+        # Spread arrival rates in main color
+        plt.plot(
+            x_values,
+            get_cyclic_data(rates2_spread),
+            marker="o",
+            color=color2,
+            label=f"{labels[1]} - Aspirational ({int(curve_params[1]*100)}% in {int(curve_params[0])}h)",
+        )
+    else:
+        # If no spread rates, use normal coloring
+        plt.plot(
+            x_values,
+            y_values2,
+            marker="o",
+            color=color2,
+            linestyle="-",
+            label=f"{labels[1]} - Arrivals",
+        )
+
+        if rates2_lagged is not None:
+            plt.plot(
+                x_values,
+                get_cyclic_data(rates2_lagged),
+                marker="s",
+                color=color2,
+                linestyle="--",
+                label=f"{labels[1]} - Lagged arrivals",
+            )
+
+    # Set plot parameters
+    plt.ylim(0, max(max(rates1), max(rates2)) + 0.25)
+    plt.xlim(hour_values[0] - x_margin, hour_values[-1] + x_margin)
+    plt.xlabel("Hour of day")
+    plt.ylabel("Arrival Rate (patients per hour)")
+    plt.title(title)
+    plt.grid(True, alpha=0.3)
+    plt.legend(bbox_to_anchor=(1.05, 1), loc='upper left')
+    plt.tight_layout()
+
+    if media_file_path:
+        filename = f"{file_prefix}{clean_title_for_filename(title)}"
+        plt.savefig(media_file_path / filename, dpi=300, bbox_inches='tight')
+
+    plt.show()