SPC/SPC_P_Chart_Instiution.py at master · Display-Lab/SPC · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from tqdm import tqdm

pd.set_option('display.max_columns', None)

#Import data
inputs = pd.read_csv("Raw_Data.csv")

#Reduce to only rows for CARD03
inputs = inputs[inputs["Measure"] == "BP01"]

#Create a row for fail rate
inputs["Fail_Rate"] = 1 - inputs["Pass_percentage"]

##Create a list of staff role
list_staff_role = inputs["Staff_Type"].drop_duplicates()

#Create empty lists to build output dataframe from
f_list_Staff_ID = []
f_list_Institution = []
f_list_Role = []
f_list_Quarters_Tracked = []
f_list_Sample_Size = []
f_list_Unwarrented_Variation = []
f_list_Magnitude_of_Variation = []
f_Rule_1_Positive = []
f_Rule_1_Magnitude = []
f_Rule_2_Positive = []
f_Rule_2_Magnitude = []
f_Rule_3_Positive = []
f_Rule_3_Magnitude = []
f_Rule_4_Positive = []
f_Rule_4_Magnitude = []
f_Rule_5_Positive = []
f_Rule_5_Magnitude = []

#Function that takes a list with a given role and returns the p-chart for each institution
for i in list_staff_role: #For each staff role
    list_institutions = inputs["Institution_ID"].drop_duplicates() #Create a list of institutions represented
    #Create a list to hold each institution's variabe
    list_i = []
    list_m = []
    list_ss =[]
    list_fr =[]
    for i in list_institutions: #For each institution
        institution_df = inputs[inputs["Institution_ID"] == i]
        list_months = institution_df["Month"].drop_duplicates()

        for m in list_months: #For each month at that institution
            institution_month_df = institution_df[institution_df["Month"] == m]
            institution = i
            month = m
            sample_size = institution_month_df["Denominator"].sum()
            fail_rate = institution_month_df["Fail_Rate"].mean()
            list_i.append(institution)
            list_m.append(month)
            list_ss.append(sample_size)
            list_fr.append(fail_rate)
    compiled_df = pd.DataFrame({ #Create new data frame that summarize each institution
        "Institution": list_i,
        "Month": list_m,
        "Sample_Size": list_ss,
        "Fail_Rate": list_fr})
    for i in list_institutions: #For each institution, generate a control chart
        compiled_institution_df = compiled_df[compiled_df["Institution"] == i]
        compiled_institution_df["Month"] = pd.to_datetime(compiled_institution_df["Month"])
        compiled_institution_df = compiled_institution_df.groupby(compiled_institution_df["Month"].dt.to_period("Q")).agg({"Sample_Size": "sum",
                                                                                                                           "Fail_Rate": "mean"}).reset_index(drop=True)
        compiled_institution_df["Institution_ID"] = i
        #compiled_institution_df = compiled_institution_df[compiled_institution_df["Sample_Size"] >= 50]
        p_bar = (compiled_institution_df["Sample_Size"] * compiled_institution_df["Fail_Rate"]).sum()/compiled_institution_df["Sample_Size"].sum()
        n_bar = compiled_institution_df["Sample_Size"].mean()
        compiled_institution_df["Standard_Deviation"] = np.sqrt((p_bar * (1 - p_bar))/(compiled_institution_df["Sample_Size"]))
        compiled_institution_df["UAL"] = (p_bar + (3 * compiled_institution_df["Standard_Deviation"]))
        compiled_institution_df["LAL"] = (p_bar - (3 * compiled_institution_df["Standard_Deviation"]))

        #plt.figure(figsize=(15,6))
        #plt.plot(compiled_institution_df["Fail_Rate"], label="Fail Rate")
        #plt.plot(compiled_institution_df["UAL"], label="UAL")
        #plt.plot(compiled_institution_df["LAL"], label="LAL")
        #plt.title(str(i))
        #plt.axhline(y=p_bar, color='blue', linestyle='-', label="Average Fail Rate")
        #leg = plt.legend()
        #plt.show()

        ##Perform SPC functions
        #Rule 1: Point above the UAL
        compiled_institution_df["R1"] = np.where((compiled_institution_df["Fail_Rate"] > compiled_institution_df["UAL"]), 1, 0)
        rule_1_m = compiled_institution_df["R1"].sum()
        rule_1_p = False
        if rule_1_m > 0:
            rule_1_p = True

        #Rule 2: 2 of 3 consecutive points above or below 2 standard deviations (Zone A or beyond)
        size = len(compiled_institution_df)
        rule_2_m = 0
        rule_2_p = False
        compiled_institution_df["R2"] = np.where((compiled_institution_df["Fail_Rate"] > compiled_institution_df["2UAL/3"]), 1, 0)
        for i in range(size - 1):
            if (compiled_institution_df["R2"][i] == 1 and compiled_institution_df["R2"][i + 1] == 1):
                rule_2_m = rule_2_m + 1
        if rule_2_m > 0:
            rule_2_p = True

        #Rule 3: 4 of 5 consecutive points above or below 1 standard deviations (Zone B or beyond)
        rule_3_m = 0
        rule_3_p = False
        compiled_institution_df["R3"] = np.where((compiled_institution_df["Fail_Rate"] > compiled_institution_df["UAL/3"]), 1, 0)
        for i in range(size - 3):
            if (compiled_institution_df["R3"][i] == 1 and compiled_institution_df["R3"][i + 1] == 1 and compiled_institution_df["R3"][i + 2] == 1 and compiled_institution_df["R3"][i + 3] == 1):
                rule_3_m = rule_3_m + 1
        if rule_3_m > 0:
            rule_3_p = True

        #Rule 4: 9 consecutive points fall on the same side of the centerline (Zone C or beyond
        rule_4_m = 0
        rule_4_p = False
        compiled_institution_df["R4"] = np.where((compiled_institution_df["Fail_Rate"] > p_bar), 1, 0)
        for i in range(size - 8):
            if (compiled_institution_df["R4"][i] == 1 and compiled_institution_df["R4"][i + 1] == 1 and compiled_institution_df["R4"][i + 2] == 1 and compiled_institution_df["R4"][i + 3] == 1 and compiled_institution_df["R4"][i + 4] == 1 and compiled_institution_df["R4"][i + 5] == 1 and compiled_institution_df["R4"][i + 6] == 1 and compiled_institution_df["R4"][i + 7] == 1 and compiled_institution_df["R4"][i + 8] == 1):
                rule_4_m = rule_4_m + 1
        if rule_4_m > 0:
            rule_4_p = True

        #Rule 5: Trend of 6 points in a row increasing or decreasing
        rule_5_m = 0
        rule_5_p = False
        for i in range(size-5):
            if (compiled_institution_df["Fail_Rate"][i] < compiled_institution_df["Fail_Rate"][i + 1]  < compiled_institution_df["Fail_Rate"][i + 2]  < compiled_institution_df["Fail_Rate"][i + 3]  < compiled_institution_df["Fail_Rate"][i + 4]  < compiled_institution_df["Fail_Rate"][i + 5]):
                rule_5_m = rule_5_m + 1
        if rule_5_m > 0:
            rule_5_p = True

        #Result cumulative of unwarrented variation
        unwarrented_variation = False
        if (rule_1_p == True or rule_2_p == True or rule_3_p == True or rule_4_p == True or rule_5_p == True):
            unwarrented_variation = True
        magnitude_variation = rule_1_m + rule_2_m + rule_3_m + rule_4_m + rule_5_m


        #Append all values to dataframe conversion lists
        f_list_Staff_ID.append(k)
        f_list_Institution.append(i)
        f_list_Role.append(j)
        f_list_Quarters_Tracked.append(len(compiled_institution_df))
        f_list_Sample_Size.append(n)
        f_list_Unwarrented_Variation.append(unwarrented_variation)
        f_list_Magnitude_of_Variation.append(magnitude_variation)
        f_Rule_1_Positive.append(rule_1_p)
        f_Rule_1_Magnitude.append(rule_1_m)
        f_Rule_2_Positive.append(rule_2_p)
        f_Rule_2_Magnitude.append(rule_2_m)
        f_Rule_3_Positive.append(rule_3_p)
        f_Rule_3_Magnitude.append(rule_3_m)
        f_Rule_4_Positive.append(rule_4_p)
        f_Rule_4_Magnitude.append(rule_4_m)
        f_Rule_5_Positive.append(rule_5_p)
        f_Rule_5_Magnitude.append(rule_5_m)

#Create a resulting dataframe
output_df = pd.DataFrame({
    "Staff_ID": f_list_Staff_ID,
    "Institution": f_list_Institution,
    "Staff_Role": f_list_Role,
    "Quarters_Tracked": f_list_Quarters_Tracked,
    "Sample_Size": f_list_Sample_Size,
    "Unwarrented_Variation_Present": f_list_Unwarrented_Variation,
    "Magnitude_of_Unwarrented_Variation":f_list_Magnitude_of_Variation,
    "Rule_1_Present": f_Rule_1_Positive,
    "Rule_1_Magnitude": f_Rule_1_Magnitude,
    "Rule_2_Present": f_Rule_2_Positive,
    "Rule_2_Magnitude": f_Rule_2_Magnitude,
    "Rule_3_Present": f_Rule_3_Positive,
    "Rule_3_Magnitude": f_Rule_3_Magnitude,
    "Rule_4_Present": f_Rule_4_Positive,
    "Rule_4_Magnitude": f_Rule_4_Magnitude,
    "Rule_5_Present": f_Rule_5_Positive,
    "Rule_5_Magnitude": f_Rule_5_Magnitude,
    })

output_df.to_csv("SPC_Results3.csv")