diff --git a/learning-projects/workshop-capacity-analysis/code.py b/learning-projects/workshop-capacity-analysis/code.py
new file mode 100644
index 0000000..ccf37b4
--- /dev/null
+++ b/learning-projects/workshop-capacity-analysis/code.py
@@ -0,0 +1,80 @@
+# %% Setup
+
+
+import random
+import pandas as pd
+import numpy as np
+import seaborn as sns
+import matplotlib.pyplot as plt
+
+
+
+def simulate_proportion_interactions_varying_formations(P, team_sizes, N_values, simulations=30):
+    """
+    Simulate team formation with varying team sizes and number of team formations,
+    and calculate the average proportion of other participants each participant interacted with.
+    """
+    experiment_data = []
+
+    for S in team_sizes:
+        for N in N_values:
+            for _ in range(simulations):
+                participants = list(range(P))
+                interactions = {p: set() for p in participants}
+
+                for _ in range(N):
+                    random.shuffle(participants)
+                    teams = [participants[i:i + S] for i in range(0, P, S)]
+                    for team in teams:
+                        for member in team:
+                            interactions[member].update(set(team) - {member})
+
+                proportions = [len(interactions[p]) / (P - 1) for p in participants]
+                
+                experiment_data.append({
+                    "team_size": S, 
+                    "N_formations": N, 
+                    "avg_proportion": np.median(proportions),
+                    "25_proportion": np.percentile(proportions, 25),
+                    "75_proportion": np.percentile(proportions, 75),
+                    "N_Participants": P
+                })
+
+    return experiment_data
+
+# %% 
+# Constants
+team_sizes = [2, 3, 4, 5]  # Team sizes
+N_values = [3, 4, 5]  # Number of team formations
+P_range = list(range(5, 31))  # Participant range from 5 to 30
+n_simulations = 400
+
+# Collecting experiment data
+all_experiment_data_varying_formations = []
+for P in P_range:
+    experiment_data = simulate_proportion_interactions_varying_formations(P, team_sizes, N_values, n_simulations)
+    all_experiment_data_varying_formations.extend(experiment_data)
+
+# %% Creating a DataFrame for the experiment data
+df = pd.DataFrame(all_experiment_data_varying_formations)
+df.to_parquet('participant_interactions_simulation.parquet')
+df
+
+# %% Plotting using seaborn
+import pandas as pd
+
+df = pd.read_parquet('participant_interactions_simulation.parquet')
+g = sns.FacetGrid(df, col="N_formations", hue="team_size", col_wrap=3, height=5)
+g.map(sns.lineplot, "N_Participants", "avg_proportion").add_legend()
+g.set_titles("Team Formations: {col_name}")
+g.set_axis_labels("Number of Participants (N)", "Average Proportion (%)")
+g.set(ylim=(0, 1), xlim=(0, max(P_range)))
+plt.show()
+
+# %% Plotting what 75% of people will at least experience
+g = sns.FacetGrid(df, col="N_formations", hue="team_size", col_wrap=3, height=5)
+g.map(sns.lineplot, "N_Participants", "25_proportion").add_legend()
+g.set_titles("Team Formations: {col_name}")
+g.set_axis_labels("Number of Participants (N)", "Average Proportion (%)")
+g.set(ylim=(0, 1), xlim=(0, max(P_range)))
+plt.show()
\ No newline at end of file
diff --git a/learning-projects/workshop-capacity-analysis/participant_interactions_simulation.parquet b/learning-projects/workshop-capacity-analysis/participant_interactions_simulation.parquet
new file mode 100644
index 0000000..fa5c389
Binary files /dev/null and b/learning-projects/workshop-capacity-analysis/participant_interactions_simulation.parquet differ
diff --git a/src/workshop-registration/attendance.py b/src/workshop-registration/attendance.py
index 56cb0f5..252a56e 100644
--- a/src/workshop-registration/attendance.py
+++ b/src/workshop-registration/attendance.py
@@ -24,8 +24,8 @@ def calculate_attendance(meeting_id):
     df = pd.DataFrame(data=data)
 
     df['date'] = df.arrived.dt.date
-    df['workshop_start'] = df.apply(lambda x: datetime.combine(x.date, time(hour= 9, minute=30)), axis = 1)
-    df['workshop_end'] = df.apply(lambda x: datetime.combine(x.date, time(hour=18, minute=00)), axis = 1)
+    df['workshop_start'] = df.apply(lambda x: datetime.combine(x.date, time(hour= 7, minute=30)), axis = 1)
+    df['workshop_end'] = df.apply(lambda x: datetime.combine(x.date, time(hour=20, minute=00)), axis = 1)
 
     df["arrival_time_corrected"] = df["arrived"].dt.tz_localize(None).clip(lower=df["workshop_start"], upper=df["workshop_end"])
     df["departure_time_corrected"] = df["departed"].dt.tz_localize(None).clip(lower=df["workshop_start"], upper=df["workshop_end"])
@@ -54,7 +54,7 @@ def calculate_attendance(meeting_id):
 
 
 if __name__ == "__main__":    
-    meeting_id = "833 3035 7443"
+    meeting_id = "87932199879"
     #meeting_id = input("Type meeting id: ")
 
     print("Calculating attendance for meeting id %s"%meeting_id)