def insert_time_from_previous(log, parameters=None):
"""
Inserts the time from the previous event, both in normal and business hours
Parameters
-------------
log
Event log
parameters
Parameters of the algorithm
Returns
-------------
enriched_log
Enriched log (with the time passed from the previous event)
"""
if parameters is None:
parameters = {}
timestamp_key = parameters[
constants.
PARAMETER_CONSTANT_TIMESTAMP_KEY] if constants.PARAMETER_CONSTANT_TIMESTAMP_KEY in parameters else xes.DEFAULT_TIMESTAMP_KEY
worktiming = parameters["worktiming"] if "worktiming" in parameters else [
7, 17
]
weekends = parameters["weekends"] if "weekends" in parameters else [6, 7]
if not type(log) is EventLog:
log = log_converter.apply(log)
log = sorting.sort_timestamp_log(log, timestamp_key)
for trace in log:
if trace:
trace[0]["@@passed_time_from_previous"] = 0
trace[0]["@@approx_bh_passed_time_from_previous"] = 0
i = 1
while i < len(trace):
trace[i]["@@passed_time_from_previous"] = (
trace[i][timestamp_key] -
trace[i - 1][timestamp_key]).total_seconds()
bh = BusinessHours(
trace[i - 1][timestamp_key].replace(tzinfo=None),
trace[i][timestamp_key].replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
trace[i][
"@@approx_bh_passed_time_from_previous"] = bh.getseconds()
i = i + 1
return log
def get_case_dispersion_avg(log, parameters=None):
"""
Gets the average time interlapsed between case ends
Parameters
--------------
log
Trace log
parameters
Parameters of the algorithm, including:
Parameters.TIMESTAMP_KEY -> attribute of the log to be used as timestamp
Returns
--------------
case_arrival_avg
Average time interlapsed between case starts
"""
if parameters is None:
parameters = {}
business_hours = exec_utils.get_param_value(Parameters.BUSINESS_HOURS,
parameters, False)
worktiming = exec_utils.get_param_value(Parameters.WORKTIMING, parameters,
[7, 17])
weekends = exec_utils.get_param_value(Parameters.WEEKENDS, parameters,
[6, 7])
timestamp_key = exec_utils.get_param_value(Parameters.TIMESTAMP_KEY,
parameters,
DEFAULT_TIMESTAMP_KEY)
case_end_time = [
trace[-1][timestamp_key] for trace in log
if trace and timestamp_key in trace[0]
]
case_end_time = sorted(case_end_time)
case_diff_end_time = []
for i in range(len(case_end_time) - 1):
if business_hours:
bh = BusinessHours(case_end_time[i].replace(tzinfo=None),
case_end_time[i + 1].replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
case_diff_end_time.append(bh.getseconds())
else:
case_diff_end_time.append(
(case_end_time[i + 1] - case_end_time[i]).total_seconds())
if case_diff_end_time:
return statistics.mean(case_diff_end_time)
return 0.0
def get_case_arrival_avg(log, parameters=None):
"""
Gets the average time interlapsed between case starts
Parameters
--------------
log
Trace log
parameters
Parameters of the algorithm, including:
PARAMETER_CONSTANT_TIMESTAMP_KEY -> attribute of the log to be used as timestamp
Returns
--------------
case_arrival_avg
Average time interlapsed between case starts
"""
if parameters is None:
parameters = {}
business_hours = parameters[
"business_hours"] if "business_hours" in parameters else False
worktiming = parameters["worktiming"] if "worktiming" in parameters else [
7, 17
]
weekends = parameters["weekends"] if "weekends" in parameters else [6, 7]
timestamp_key = parameters[
PARAMETER_CONSTANT_TIMESTAMP_KEY] if PARAMETER_CONSTANT_TIMESTAMP_KEY in parameters else DEFAULT_TIMESTAMP_KEY
case_start_time = [
trace[0][timestamp_key] for trace in log
if trace and timestamp_key in trace[0]
]
case_start_time = sorted(case_start_time)
case_diff_start_time = []
for i in range(len(case_start_time) - 1):
if business_hours:
bh = BusinessHours(case_start_time[i].replace(tzinfo=None),
case_start_time[i + 1].replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
case_diff_start_time.append(bh.getseconds())
else:
case_diff_start_time.append(
(case_start_time[i + 1] - case_start_time[i]).total_seconds())
if case_diff_start_time:
return statistics.median(case_diff_start_time)
return 0.0
def get_remaining_time_from_log(log, max_len_trace=100000, parameters=None):
"""
Gets the remaining time for the instances given a log and a trace index
Parameters
------------
log
Log
max_len_trace
Index
parameters
Parameters of the algorithm
Returns
------------
list
List of remaining times
"""
if parameters is None:
parameters = {}
timestamp_key = parameters[
constants.
PARAMETER_CONSTANT_TIMESTAMP_KEY] if constants.PARAMETER_CONSTANT_TIMESTAMP_KEY in parameters else xes.DEFAULT_TIMESTAMP_KEY
business_hours = parameters[
"business_hours"] if "business_hours" in parameters else False
worktiming = parameters["worktiming"] if "worktiming" in parameters else [
7, 17
]
weekends = parameters["weekends"] if "weekends" in parameters else [6, 7]
y_orig = []
for trace in log:
y_orig.append([])
for index, event in enumerate(trace):
if index >= max_len_trace:
break
timestamp_st = trace[index][timestamp_key]
timestamp_et = trace[-1][timestamp_key]
if business_hours:
bh = BusinessHours(timestamp_st.replace(tzinfo=None),
timestamp_et.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
y_orig[-1].append(bh.getseconds())
else:
y_orig[-1].append(
(timestamp_et - timestamp_st).total_seconds())
while len(y_orig[-1]) < max_len_trace:
y_orig[-1].append(y_orig[-1][-1])
return y_orig
def execute_script():
ts1 = 100000000
ts2 = 110000000
d1 = datetime.datetime.fromtimestamp(ts1)
d2 = datetime.datetime.fromtimestamp(ts2)
print(ts2 - ts1)
# default business hours: all the days of the week except Saturday and Sunday are working days.
bh1 = BusinessHours(d1, d2, worktiming=[[7, 12.5], [13, 17]])
print(bh1.getseconds())
# let's calculate the business hours using a proper work calendar.
bh2 = BusinessHours(d1,
d2,
worktiming=[[7, 12.25], [13.25, 17]],
workcalendar=Italy())
print(bh2.getseconds())
def get_case_arrival_avg(
log: EventLog,
parameters: Optional[Dict[Union[str, Parameters],
Any]] = None) -> float:
"""
Gets the average time interlapsed between case starts
Parameters
--------------
log
Trace log
parameters
Parameters of the algorithm, including:
Parameters.TIMESTAMP_KEY -> attribute of the log to be used as timestamp
Returns
--------------
case_arrival_avg
Average time interlapsed between case starts
"""
if parameters is None:
parameters = {}
log = log_converter.apply(log,
variant=log_converter.Variants.TO_EVENT_LOG,
parameters=parameters)
business_hours = exec_utils.get_param_value(Parameters.BUSINESS_HOURS,
parameters, False)
worktiming = exec_utils.get_param_value(Parameters.WORKTIMING, parameters,
[7, 17])
weekends = exec_utils.get_param_value(Parameters.WEEKENDS, parameters,
[6, 7])
workcalendar = exec_utils.get_param_value(
Parameters.WORKCALENDAR, parameters,
constants.DEFAULT_BUSINESS_HOURS_WORKCALENDAR)
timestamp_key = exec_utils.get_param_value(Parameters.TIMESTAMP_KEY,
parameters,
DEFAULT_TIMESTAMP_KEY)
case_start_time = [
trace[0][timestamp_key] for trace in log
if trace and timestamp_key in trace[0]
]
case_start_time = sorted(case_start_time)
case_diff_start_time = []
for i in range(len(case_start_time) - 1):
if business_hours:
bh = BusinessHours(case_start_time[i].replace(tzinfo=None),
case_start_time[i + 1].replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends,
workcalendar=workcalendar)
case_diff_start_time.append(bh.getseconds())
else:
case_diff_start_time.append(
(case_start_time[i + 1] - case_start_time[i]).total_seconds())
if case_diff_start_time:
return statistics.mean(case_diff_start_time)
return 0.0
def performance(
log: Union[EventLog, EventStream],
parameters: Optional[Dict[Union[str, Parameters], Any]] = None
) -> Dict[Tuple[str, str], float]:
"""
Measure performance between couples of attributes in the DFG graph
Parameters
----------
log
Log
parameters
Possible parameters passed to the algorithms:
aggregationMeasure -> performance aggregation measure (min, max, mean, median)
activity_key -> Attribute to use as activity
timestamp_key -> Attribute to use as timestamp
- Parameters.BUSINESS_HOURS => calculates the difference of time based on the business hours, not the total time.
Default: False
- Parameters.WORKTIMING => work schedule of the company (provided as a list where the first number is the start
of the work time, and the second number is the end of the work time), if business hours are enabled
Default: [7, 17] (work shift from 07:00 to 17:00)
- Parameters.WEEKENDS => indexes of the days of the week that are weekend
Default: [6, 7] (weekends are Saturday and Sunday)
Returns
-------
dfg
DFG graph
"""
if parameters is None:
parameters = {}
activity_key = exec_utils.get_param_value(Parameters.ACTIVITY_KEY,
parameters,
xes_util.DEFAULT_NAME_KEY)
start_timestamp_key = exec_utils.get_param_value(
Parameters.START_TIMESTAMP_KEY, parameters,
xes_util.DEFAULT_TIMESTAMP_KEY)
timestamp_key = exec_utils.get_param_value(Parameters.TIMESTAMP_KEY,
parameters,
xes_util.DEFAULT_TIMESTAMP_KEY)
aggregation_measure = exec_utils.get_param_value(
Parameters.AGGREGATION_MEASURE, parameters, "mean")
business_hours = exec_utils.get_param_value(Parameters.BUSINESS_HOURS,
parameters, False)
worktiming = exec_utils.get_param_value(Parameters.WORKTIMING, parameters,
[7, 17])
weekends = exec_utils.get_param_value(Parameters.WEEKENDS, parameters,
[6, 7])
workcalendar = exec_utils.get_param_value(
Parameters.WORKCALENDAR, parameters,
constants.DEFAULT_BUSINESS_HOURS_WORKCALENDAR)
if business_hours:
dfgs0 = map((lambda t: [(
(t[i - 1][activity_key], t[i][activity_key]),
max(
0,
BusinessHours(t[i - 1][timestamp_key].replace(tzinfo=None),
t[i][start_timestamp_key].replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends,
workcalendar=workcalendar).getseconds()))
for i in range(1, len(t))]), log)
else:
dfgs0 = map((lambda t: [((t[i - 1][activity_key], t[i][activity_key]),
max(0, (t[i][start_timestamp_key] - t[i - 1][
timestamp_key]).total_seconds()))
for i in range(1, len(t))]), log)
ret0 = {}
for el in dfgs0:
for couple in el:
if not couple[0] in ret0:
ret0[couple[0]] = []
ret0[couple[0]].append(couple[1])
ret = Counter()
for key in ret0:
if aggregation_measure == "median":
ret[key] = median(ret0[key])
elif aggregation_measure == "min":
ret[key] = min(ret0[key])
elif aggregation_measure == "max":
ret[key] = max(ret0[key])
elif aggregation_measure == "stdev":
ret[key] = stdev(ret0[key]) if len(ret0[key]) > 1 else 0
elif aggregation_measure == "sum":
ret[key] = sum(ret0[key])
elif aggregation_measure == "raw_values":
ret[key] = ret0[key]
elif aggregation_measure == "all":
ret[key] = {
"median": median(ret0[key]),
"min": min(ret0[key]),
"max": max(ret0[key]),
"stdev": stdev(ret0[key]) if len(ret0[key]) > 1 else 0,
"sum": sum(ret0[key]),
"mean": mean(ret0[key])
}
else:
ret[key] = mean(ret0[key])
return ret
def to_interval(log, parameters=None):
"""
Converts a log to interval format (e.g. an event has two timestamps)
from lifecycle format (an event has only a timestamp, and a transition lifecycle)
Parameters
-------------
log
Log (expressed in the lifecycle format)
parameters
Possible parameters of the method (activity, timestamp key, start timestamp key, transition ...)
Returns
-------------
log
Interval event log
"""
if parameters is None:
parameters = {}
timestamp_key = parameters[
constants.
PARAMETER_CONSTANT_TIMESTAMP_KEY] if constants.PARAMETER_CONSTANT_TIMESTAMP_KEY in parameters else xes.DEFAULT_TIMESTAMP_KEY
start_timestamp_key = parameters[
constants.
PARAMETER_CONSTANT_START_TIMESTAMP_KEY] if constants.PARAMETER_CONSTANT_START_TIMESTAMP_KEY in parameters else xes.DEFAULT_START_TIMESTAMP_KEY
transition_key = parameters[
constants.
PARAMETER_CONSTANT_TRANSITION_KEY] if constants.PARAMETER_CONSTANT_TRANSITION_KEY in parameters else xes.DEFAULT_TRANSITION_KEY
activity_key = parameters[
constants.
PARAMETER_CONSTANT_ACTIVITY_KEY] if constants.PARAMETER_CONSTANT_ACTIVITY_KEY in parameters else xes.DEFAULT_NAME_KEY
business_hours = parameters[
"business_hours"] if "business_hours" in parameters else False
worktiming = parameters["worktiming"] if "worktiming" in parameters else [
7, 17
]
weekends = parameters["weekends"] if "weekends" in parameters else [6, 7]
if log is not None and len(log) > 0:
if "PM4PY_TYPE" in log.attributes and log.attributes[
"PM4PY_TYPE"] == "interval":
return log
if log[0] is not None and len(log[0]) > 0:
first_event = log[0][0]
if start_timestamp_key in first_event:
return log
new_log = EventLog()
new_log.attributes["PM4PY_TYPE"] = "interval"
for trace in log:
new_trace = Trace()
for attr in trace.attributes:
new_trace.attributes[attr] = trace.attributes[attr]
activities_start = {}
for event in trace:
activity = event[activity_key]
transition = event[
transition_key] if transition_key in event else "complete"
timestamp = event[timestamp_key]
if transition.lower() == "start":
if activity not in activities_start:
activities_start[activity] = list()
activities_start[activity].append(event)
elif transition.lower() == "complete":
start_event = None
start_timestamp = event[timestamp_key]
if activity in activities_start and len(
activities_start[activity]) > 0:
start_event = activities_start[activity].pop(0)
start_timestamp = start_event[timestamp_key]
new_event = Event()
for attr in event:
if not attr == timestamp_key and not attr == transition_key:
new_event[attr] = event[attr]
if start_event is not None:
for attr in start_event:
if not attr == timestamp_key and not attr == transition_key:
new_event["@@startevent_" +
attr] = start_event[attr]
new_event[start_timestamp_key] = start_timestamp
new_event[timestamp_key] = timestamp
new_event["@@duration"] = (
timestamp - start_timestamp).total_seconds()
if business_hours:
bh = BusinessHours(
start_timestamp.replace(tzinfo=None),
timestamp.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
new_event["@@approx_bh_duration"] = bh.getseconds()
new_trace.append(new_event)
new_trace = sorting.sort_timestamp_trace(new_trace,
start_timestamp_key)
new_log.append(new_trace)
return new_log
return log
def assign_lead_cycle_time(log, parameters=None):
"""
Assigns the lead and cycle time to an interval log
Parameters
-------------
log
Interval log
parameters
Parameters of the algorithm, including: start_timestamp_key, timestamp_key, worktiming, weekends
"""
if parameters is None:
parameters = {}
start_timestamp_key = parameters[
constants.
PARAMETER_CONSTANT_START_TIMESTAMP_KEY] if constants.PARAMETER_CONSTANT_START_TIMESTAMP_KEY in parameters else xes.DEFAULT_START_TIMESTAMP_KEY
timestamp_key = parameters[
constants.
PARAMETER_CONSTANT_TIMESTAMP_KEY] if constants.PARAMETER_CONSTANT_TIMESTAMP_KEY in parameters else xes.DEFAULT_TIMESTAMP_KEY
worktiming = parameters["worktiming"] if "worktiming" in parameters else [
7, 17
]
weekends = parameters["weekends"] if "weekends" in parameters else [6, 7]
interval_log = to_interval(log, parameters=parameters)
for trace in interval_log:
approx_partial_lead_time = 0
approx_partial_cycle_time = 0
approx_wasted_time = 0
max_et = None
max_et_seconds = 0
for i in range(len(trace)):
this_wasted_time = 0
st = trace[i][start_timestamp_key]
st_seconds = st.timestamp()
et = trace[i][timestamp_key]
et_seconds = et.timestamp()
if max_et_seconds > 0 and st_seconds > max_et_seconds:
bh_unworked = BusinessHours(max_et.replace(tzinfo=None),
st.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
unworked_sec = bh_unworked.getseconds()
approx_partial_lead_time = approx_partial_lead_time + unworked_sec
approx_wasted_time = approx_wasted_time + unworked_sec
this_wasted_time = unworked_sec
if st_seconds > max_et_seconds:
bh = BusinessHours(st.replace(tzinfo=None),
et.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
approx_bh_duration = bh.getseconds()
approx_partial_cycle_time = approx_partial_cycle_time + approx_bh_duration
approx_partial_lead_time = approx_partial_lead_time + approx_bh_duration
elif st_seconds < max_et_seconds and et_seconds > max_et_seconds:
bh = BusinessHours(max_et.replace(tzinfo=None),
et.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
approx_bh_duration = bh.getseconds()
approx_partial_cycle_time = approx_partial_cycle_time + approx_bh_duration
approx_partial_lead_time = approx_partial_lead_time + approx_bh_duration
if et_seconds > max_et_seconds:
max_et_seconds = et_seconds
max_et = et
ratio_cycle_lead_time = 1
if approx_partial_lead_time > 0:
ratio_cycle_lead_time = approx_partial_cycle_time / approx_partial_lead_time
trace[i][
"@@approx_bh_partial_cycle_time"] = approx_partial_cycle_time
trace[i][
"@@approx_bh_partial_lead_time"] = approx_partial_lead_time
trace[i]["@@approx_bh_overall_wasted_time"] = approx_wasted_time
trace[i]["@@approx_bh_this_wasted_time"] = this_wasted_time
trace[i][
"@approx_bh_ratio_cycle_lead_time"] = ratio_cycle_lead_time
return interval_log
case_dispersion_ratio = case_arrival.get_case_dispersion_avg(log, parameters={ case_arrival.Parameters.TIMESTAMP_KEY: "time:timestamp"})
case_dispersion_ratio
#Peformance Spectrum
from pm4py.statistics.performance_spectrum import algorithm as performance_spectrum
ps = performance_spectrum.apply(log, ["register request", "decide"], parameters= {performance_spectrum.Parameters.ACTIVITY_KEY: "concept:name", performance_spectrum.Parameters.TIMESTAMP_KEY: "time:timestamp"})
ps
#Business Hours
from pm4py.util.business_hours import BusinessHours
from datetime import datetime
st = datetime.fromtimestamp(100000000)
et = datetime.fromtimestamp(200000000)
bh_object = BusinessHours(st, et)
worked_time = bh_object.getseconds()
print(worked_time)
#Cycle Time and Waiting Time
from pm4py.objects.log.util import interval_lifecycle
enriched_log = interval_lifecycle.assign_lead_cycle_time(log)
#error
#display graphs
import os
from pm4py.objects.log.importer.xes import importer as xes_importer
log_path = os.path.join("tests","input_data","receipt.xes")
log = xes_importer.apply(log_path)
def train(log, parameters=None):
"""
Train the prediction model
Parameters
-----------
log
Event log
parameters
Possible parameters of the algorithm
Returns
------------
model
Trained model
"""
if parameters is None:
parameters = {}
parameters["enable_sort"] = False
activity_key = parameters[
constants.
PARAMETER_CONSTANT_ACTIVITY_KEY] if constants.PARAMETER_CONSTANT_ACTIVITY_KEY in parameters else xes.DEFAULT_NAME_KEY
timestamp_key = parameters[
constants.
PARAMETER_CONSTANT_TIMESTAMP_KEY] if constants.PARAMETER_CONSTANT_TIMESTAMP_KEY in parameters else xes.DEFAULT_TIMESTAMP_KEY
business_hours = parameters[
"business_hours"] if "business_hours" in parameters else False
worktiming = parameters["worktiming"] if "worktiming" in parameters else [
7, 17
]
weekends = parameters["weekends"] if "weekends" in parameters else [6, 7]
y_orig = parameters["y_orig"] if "y_orig" in parameters else None
log = sorting.sort_timestamp(log, timestamp_key)
str_evsucc_attr = [activity_key]
if "str_ev_attr" in parameters:
str_tr_attr = parameters[
"str_tr_attr"] if "str_tr_attr" in parameters else []
str_ev_attr = parameters[
"str_ev_attr"] if "str_ev_attr" in parameters else []
num_tr_attr = parameters[
"num_tr_attr"] if "num_tr_attr" in parameters else []
num_ev_attr = parameters[
"num_ev_attr"] if "num_ev_attr" in parameters else []
else:
str_tr_attr, str_ev_attr, num_tr_attr, num_ev_attr = attributes_filter.select_attributes_from_log_for_tree(
log)
if activity_key not in str_ev_attr:
str_ev_attr.append(activity_key)
max_trace_length = max(len(x) for x in log)
if max_trace_length == 1:
# this you shall use
data, feature_names = get_log_representation.get_representation(
log,
str_tr_attr,
str_ev_attr,
num_tr_attr,
num_ev_attr,
str_evsucc_attr=str_evsucc_attr)
ext_log = log
else:
ext_log, change_indexes = get_log_with_log_prefixes(log)
data, feature_names = get_log_representation.get_representation(
ext_log,
str_tr_attr,
str_ev_attr,
num_tr_attr,
num_ev_attr,
str_evsucc_attr=str_evsucc_attr)
if y_orig is not None:
remaining_time = [y for x in y_orig for y in x]
else:
if business_hours:
remaining_time = []
for trace in ext_log:
if trace:
timestamp_et = trace[-1][timestamp_key]
timestamp_st = trace[0][timestamp_key]
bh = BusinessHours(timestamp_st.replace(tzinfo=None),
timestamp_et.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
remaining_time.append(bh.getseconds())
else:
remaining_time.append(0)
else:
remaining_time = []
for trace in ext_log:
if trace:
remaining_time.append(
(trace[-1][timestamp_key] -
trace[0][timestamp_key]).total_seconds())
else:
remaining_time.append(0)
regr = ElasticNet(max_iter=10000, l1_ratio=0.7)
print(data)
regr.fit(data, remaining_time)
return {
"str_tr_attr": str_tr_attr,
"str_ev_attr": str_ev_attr,
"num_tr_attr": num_tr_attr,
"num_ev_attr": num_ev_attr,
"str_evsucc_attr": str_evsucc_attr,
"feature_names": feature_names,
"remaining_time": remaining_time,
"regr": regr,
"variant": "elasticnet"
}
def to_interval(log, parameters=None):
"""
Converts a log to interval format (e.g. an event has two timestamps)
from lifecycle format (an event has only a timestamp, and a transition lifecycle)
Parameters
-------------
log
Log (expressed in the lifecycle format)
parameters
Possible parameters of the method (activity, timestamp key, start timestamp key, transition ...)
Returns
-------------
log
Interval event log
"""
if parameters is None:
parameters = {}
timestamp_key = exec_utils.get_param_value(Parameters.TIMESTAMP_KEY,
parameters,
xes.DEFAULT_TIMESTAMP_KEY)
start_timestamp_key = exec_utils.get_param_value(
Parameters.START_TIMESTAMP_KEY, parameters,
xes.DEFAULT_START_TIMESTAMP_KEY)
transition_key = exec_utils.get_param_value(Parameters.TRANSITION_KEY,
parameters,
xes.DEFAULT_TRANSITION_KEY)
activity_key = exec_utils.get_param_value(Parameters.ACTIVITY_KEY,
parameters, xes.DEFAULT_NAME_KEY)
lifecycle_instance_key = exec_utils.get_param_value(
Parameters.LIFECYCLE_INSTANCE_KEY, parameters,
xes.DEFAULT_INSTANCE_KEY)
business_hours = exec_utils.get_param_value(Parameters.BUSINESS_HOURS,
parameters, False)
worktiming = exec_utils.get_param_value(Parameters.WORKTIMING, parameters,
[7, 17])
weekends = exec_utils.get_param_value(Parameters.WEEKENDS, parameters,
[6, 7])
if log is not None and len(log) > 0:
if "PM4PY_TYPE" in log.attributes and log.attributes[
"PM4PY_TYPE"] == "interval":
return log
if log[0] is not None and len(log[0]) > 0:
first_event = log[0][0]
if start_timestamp_key in first_event:
return log
new_log = EventLog(attributes=copy(log.attributes),
extensions=copy(log.extensions),
classifiers=copy(log.classifiers),
omni_present=copy(log.omni_present),
properties=copy(log.properties))
new_log.attributes["PM4PY_TYPE"] = "interval"
new_log.properties[
constants.
PARAMETER_CONSTANT_START_TIMESTAMP_KEY] = xes.DEFAULT_START_TIMESTAMP_KEY
for trace in log:
new_trace = Trace()
for attr in trace.attributes:
new_trace.attributes[attr] = trace.attributes[attr]
activities_start = {}
for event in trace:
activity = event[activity_key]
instance = event[
lifecycle_instance_key] if lifecycle_instance_key in event else None
activity = (activity, instance)
transition = event[
transition_key] if transition_key in event else "complete"
timestamp = event[timestamp_key]
if transition.lower() == "start":
if activity not in activities_start:
activities_start[activity] = list()
activities_start[activity].append(event)
elif transition.lower() == "complete":
start_event = None
start_timestamp = event[timestamp_key]
if activity in activities_start and len(
activities_start[activity]) > 0:
start_event = activities_start[activity].pop(0)
start_timestamp = start_event[timestamp_key]
new_event = Event()
for attr in event:
if not attr == timestamp_key and not attr == transition_key:
new_event[attr] = event[attr]
if start_event is not None:
for attr in start_event:
if not attr == timestamp_key and not attr == transition_key:
new_event["@@startevent_" +
attr] = start_event[attr]
new_event[start_timestamp_key] = start_timestamp
new_event[timestamp_key] = timestamp
new_event["@@duration"] = (
timestamp - start_timestamp).total_seconds()
if business_hours:
bh = BusinessHours(
start_timestamp.replace(tzinfo=None),
timestamp.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
new_event["@@approx_bh_duration"] = bh.getseconds()
new_trace.append(new_event)
new_trace = sorting.sort_timestamp_trace(new_trace,
start_timestamp_key)
new_log.append(new_trace)
return new_log
return log
def assign_lead_cycle_time(log, parameters=None):
"""
Assigns the lead and cycle time to an interval log
Parameters
-------------
log
Interval log
parameters
Parameters of the algorithm, including: start_timestamp_key, timestamp_key, worktiming, weekends
"""
if parameters is None:
parameters = {}
timestamp_key = exec_utils.get_param_value(Parameters.TIMESTAMP_KEY,
parameters,
xes.DEFAULT_TIMESTAMP_KEY)
start_timestamp_key = exec_utils.get_param_value(
Parameters.START_TIMESTAMP_KEY, parameters,
xes.DEFAULT_START_TIMESTAMP_KEY)
worktiming = exec_utils.get_param_value(Parameters.WORKTIMING, parameters,
[7, 17])
weekends = exec_utils.get_param_value(Parameters.WEEKENDS, parameters,
[6, 7])
interval_log = to_interval(log, parameters=parameters)
for trace in interval_log:
approx_partial_lead_time = 0
approx_partial_cycle_time = 0
approx_wasted_time = 0
max_et = None
max_et_seconds = 0
for i in range(len(trace)):
this_wasted_time = 0
st = trace[i][start_timestamp_key]
st_seconds = st.timestamp()
et = trace[i][timestamp_key]
et_seconds = et.timestamp()
if max_et_seconds > 0 and st_seconds > max_et_seconds:
bh_unworked = BusinessHours(max_et.replace(tzinfo=None),
st.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
unworked_sec = bh_unworked.getseconds()
approx_partial_lead_time = approx_partial_lead_time + unworked_sec
approx_wasted_time = approx_wasted_time + unworked_sec
this_wasted_time = unworked_sec
if st_seconds > max_et_seconds:
bh = BusinessHours(st.replace(tzinfo=None),
et.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
approx_bh_duration = bh.getseconds()
approx_partial_cycle_time = approx_partial_cycle_time + approx_bh_duration
approx_partial_lead_time = approx_partial_lead_time + approx_bh_duration
elif st_seconds < max_et_seconds and et_seconds > max_et_seconds:
bh = BusinessHours(max_et.replace(tzinfo=None),
et.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
approx_bh_duration = bh.getseconds()
approx_partial_cycle_time = approx_partial_cycle_time + approx_bh_duration
approx_partial_lead_time = approx_partial_lead_time + approx_bh_duration
if et_seconds > max_et_seconds:
max_et_seconds = et_seconds
max_et = et
ratio_cycle_lead_time = 1
if approx_partial_lead_time > 0:
ratio_cycle_lead_time = approx_partial_cycle_time / approx_partial_lead_time
trace[i][
"@@approx_bh_partial_cycle_time"] = approx_partial_cycle_time
trace[i][
"@@approx_bh_partial_lead_time"] = approx_partial_lead_time
trace[i]["@@approx_bh_overall_wasted_time"] = approx_wasted_time
trace[i]["@@approx_bh_this_wasted_time"] = this_wasted_time
trace[i][
"@approx_bh_ratio_cycle_lead_time"] = ratio_cycle_lead_time
return interval_log
def apply(
log: EventLog,
parameters: Optional[Dict[Any, Any]] = None) -> typing.TemporalProfile:
"""
Gets the temporal profile from the log.
Implements the approach described in:
Stertz, Florian, Jürgen Mangler, and Stefanie Rinderle-Ma. "Temporal Conformance Checking at Runtime based on Time-infused Process Models." arXiv preprint arXiv:2008.07262 (2020).
Parameters
----------
log
Event log
parameters
Parameters, including:
- Parameters.ACTIVITY_KEY => the attribute to use as activity
- Parameters.START_TIMESTAMP_KEY => the attribute to use as start timestamp
- Parameters.TIMESTAMP_KEY => the attribute to use as timestamp
- Parameters.BUSINESS_HOURS => calculates the difference of time based on the business hours, not the total time.
Default: False
- Parameters.WORKTIMING => work schedule of the company (provided as a list where the first number is the start
of the work time, and the second number is the end of the work time), if business hours are enabled
Default: [7, 17] (work shift from 07:00 to 17:00)
- Parameters.WEEKENDS => indexes of the days of the week that are weekend
Default: [6, 7] (weekends are Saturday and Sunday)
Returns
-------
temporal_profile
Temporal profile of the log
"""
if parameters is None:
parameters = {}
log = log_converter.apply(log,
variant=log_converter.Variants.TO_EVENT_LOG,
parameters=parameters)
business_hours = exec_utils.get_param_value(Parameters.BUSINESS_HOURS,
parameters, False)
worktiming = exec_utils.get_param_value(Parameters.WORKTIMING, parameters,
[7, 17])
weekends = exec_utils.get_param_value(Parameters.WEEKENDS, parameters,
[6, 7])
workcalendar = exec_utils.get_param_value(
Parameters.WORKCALENDAR, parameters,
constants.DEFAULT_BUSINESS_HOURS_WORKCALENDAR)
activity_key = exec_utils.get_param_value(Parameters.ACTIVITY_KEY,
parameters,
xes_constants.DEFAULT_NAME_KEY)
timestamp_key = exec_utils.get_param_value(
Parameters.TIMESTAMP_KEY, parameters,
xes_constants.DEFAULT_TIMESTAMP_KEY)
start_timestamp_key = exec_utils.get_param_value(
Parameters.START_TIMESTAMP_KEY, parameters,
xes_constants.DEFAULT_TIMESTAMP_KEY)
diff_time_recordings = {}
for trace in log:
for i in range(len(trace) - 1):
act_i = trace[i][activity_key]
time_i = trace[i][timestamp_key].timestamp()
for j in range(i + 1, len(trace)):
time_j = trace[j][start_timestamp_key].timestamp()
if time_j >= time_i:
act_j = trace[j][activity_key]
if not (act_i, act_j) in diff_time_recordings:
diff_time_recordings[(act_i, act_j)] = []
if business_hours:
bh = BusinessHours(
trace[i][timestamp_key].replace(tzinfo=None),
trace[j][start_timestamp_key].replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends,
workcalendar=workcalendar)
diff_time_recordings[(act_i,
act_j)].append(bh.getseconds())
else:
diff_time_recordings[(act_i,
act_j)].append(time_j - time_i)
temporal_profile = {}
for ac in diff_time_recordings:
if len(diff_time_recordings[ac]) > 1:
temporal_profile[ac] = (mean(diff_time_recordings[ac]),
stdev(diff_time_recordings[ac]))
else:
temporal_profile[ac] = (diff_time_recordings[ac][0], 0)
return temporal_profile
def apply(
log: EventLog,
parameters: Optional[Dict[Union[str, Parameters], Any]] = None
) -> Dict[str, float]:
"""
Gets the sojourn time per activity on an event log object
Parameters
--------------
dataframe
Pandas dataframe
parameters
Parameters of the algorithm, including:
- Parameters.ACTIVITY_KEY => activity key
- Parameters.START_TIMESTAMP_KEY => start timestamp key
- Parameters.TIMESTAMP_KEY => timestamp key
- Parameters.BUSINESS_HOURS => calculates the difference of time based on the business hours, not the total time.
Default: False
- Parameters.WORKTIMING => work schedule of the company (provided as a list where the first number is the start
of the work time, and the second number is the end of the work time), if business hours are enabled
Default: [7, 17] (work shift from 07:00 to 17:00)
- Parameters.WEEKENDS => indexes of the days of the week that are weekend
Default: [6, 7] (weekends are Saturday and Sunday)
- Parameters.AGGREGATION_MEASURE => performance aggregation measure (sum, min, max, mean, median)
Returns
--------------
soj_time_dict
Sojourn time dictionary
"""
if parameters is None:
parameters = {}
business_hours = exec_utils.get_param_value(Parameters.BUSINESS_HOURS,
parameters, False)
worktiming = exec_utils.get_param_value(Parameters.WORKTIMING, parameters,
[7, 17])
weekends = exec_utils.get_param_value(Parameters.WEEKENDS, parameters,
[6, 7])
workcalendar = exec_utils.get_param_value(
Parameters.WORKCALENDAR, parameters,
constants.DEFAULT_BUSINESS_HOURS_WORKCALENDAR)
log = log_converter.apply(log,
variant=log_converter.Variants.TO_EVENT_LOG,
parameters=parameters)
activity_key = exec_utils.get_param_value(Parameters.ACTIVITY_KEY,
parameters,
xes_constants.DEFAULT_NAME_KEY)
start_timestamp_key = exec_utils.get_param_value(
Parameters.START_TIMESTAMP_KEY, parameters,
xes_constants.DEFAULT_TIMESTAMP_KEY)
timestamp_key = exec_utils.get_param_value(
Parameters.TIMESTAMP_KEY, parameters,
xes_constants.DEFAULT_TIMESTAMP_KEY)
aggregation_measure = exec_utils.get_param_value(
Parameters.AGGREGATION_MEASURE, parameters, "mean")
durations_dict = {}
activities = [ev[activity_key] for trace in log for ev in trace]
for act in activities:
durations_dict[act] = []
for trace in log:
for event in trace:
activity = event[activity_key]
if business_hours:
bh = BusinessHours(
event[start_timestamp_key].replace(tzinfo=None),
event[timestamp_key].replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends,
workcalendar=workcalendar)
durations_dict[activity].append(bh.getseconds())
else:
start_time = event[start_timestamp_key].timestamp()
complete_time = event[timestamp_key].timestamp()
durations_dict[activity].append(complete_time - start_time)
for act in durations_dict:
if aggregation_measure == "median":
durations_dict[act] = median(durations_dict[act])
elif aggregation_measure == "min":
durations_dict[act] = min(durations_dict[act])
elif aggregation_measure == "max":
durations_dict[act] = max(durations_dict[act])
elif aggregation_measure == "sum":
durations_dict[act] = sum(durations_dict[act])
else:
durations_dict[act] = mean(durations_dict[act])
return durations_dict
def get_class_representation_by_trace_duration(log,
target_trace_duration,
timestamp_key="time:timestamp",
parameters=None):
"""
Get class representation by splitting traces according to trace duration
Parameters
------------
log
Trace log
target_trace_duration
Target trace duration
timestamp_key
Timestamp key
Returns
------------
target
Target part for decision tree learning
classes
Name of the classes, in order
"""
if parameters is None:
parameters = {}
business_hours = parameters[
"business_hours"] if "business_hours" in parameters else False
worktiming = parameters["worktiming"] if "worktiming" in parameters else [
7, 17
]
weekends = parameters["weekends"] if "weekends" in parameters else [6, 7]
count = 0
dictionary = {}
target = []
classes = []
for trace in log:
value = "LESSEQUAL"
if len(trace) > 0 and timestamp_key in trace[
0] and timestamp_key in trace[-1]:
timestamp_st = trace[0][timestamp_key]
timestamp_et = trace[-1][timestamp_key]
if business_hours:
bh = BusinessHours(timestamp_st.replace(tzinfo=None),
timestamp_et.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
diff = bh.getseconds()
else:
diff = (timestamp_et - timestamp_st).total_seconds()
if diff > target_trace_duration:
value = "GREATER"
if not str(value) in dictionary:
dictionary[str(value)] = count
classes.append(str(value))
count = count + 1
target.append(dictionary[str(value)])
target = np.array(target)
return target, classes
def statics(request):
logadr = ADRESS
log = infra.recieve_and_convert_log.convert_log(logadr, logname, logtime,
logtran, logstart,
logcompl, logreso, logid,
inputname[-3:])
ptree = infra.recieve_and_convert_log.get_processtree(log)
#duration = infra.recieve_and_convert_log.get_duration(log)
duration = Duration
#deviation = infra.recieve_and_convert_log.get_deviation(duration,log)
deviation = Deviation
#waitingtime = infra.recieve_and_convert_log.waitingtime(log)
waitingtime = Waitingtime
#frequency = infra.recieve_and_convert_log.get_waitinhour(log,Waitingtime,'n',Watichange)
#frequency = Frequency
st = datetime.fromtimestamp(100000000)
et = datetime.fromtimestamp(200000000)
bh_object = BusinessHours(st, et)
worked_time = bh_object.getseconds()
#log_path = os.path.join("tests","input_data","receipt.xes")
initialtrace = infra.recieve_and_convert_log.initialtrace(log)
x, y = case_statistics.get_kde_caseduration(
log,
parameters={
constants.PARAMETER_CONSTANT_TIMESTAMP_KEY: "time:timestamp"
})
gviz1 = graphs_visualizer.apply_plot(
x, y, variant=graphs_visualizer.Variants.CASES)
gviz2 = graphs_visualizer.apply_semilogx(
x, y, variant=graphs_visualizer.Variants.CASES)
graphs_visualizer.save(gviz1, "DES1/static/image1.gv.png")
graphs_visualizer.save(gviz2, "DES1/static/image2.gv.png")
x, y = attributes_filter.get_kde_date_attribute(log,
attribute="time:timestamp")
gviz3 = graphs_visualizer.apply_plot(
x, y, variant=graphs_visualizer.Variants.DATES)
graphs_visualizer.save(gviz3, "DES1/static/image3.gv.png")
'''
x, y = attributes_filter.get_kde_numeric_attribute(log, "amount")
gviz4 = graphs_visualizer.apply_plot(x, y, variant=graphs_visualizer.Variants.ATTRIBUTES)
gviz5 = graphs_visualizer.apply_semilogx(x, y, variant=graphs_visualizer.Variants.ATTRIBUTES)
graphs_visualizer.save(gviz4,"./static/image4.gv.png")
graphs_visualizer.save(gviz5,"./static/image5.gv.png")
'''
numtrace = infra.recieve_and_convert_log.statics(log)[0]
numactivity = infra.recieve_and_convert_log.statics(log)[1]
activitylist = infra.recieve_and_convert_log.statics(log)[2]
timeinterval = infra.recieve_and_convert_log.statics(log)[3]
meanthoughputtime = infra.recieve_and_convert_log.statics(log)[4][0]
deviationthoughputtime = infra.recieve_and_convert_log.statics(log)[4][1]
arrivalratio = infra.recieve_and_convert_log.statics(log)[5]
dispersionratio = infra.recieve_and_convert_log.statics(log)[6]
resourcedict = infra.recieve_and_convert_log.initialresource1(log)
initialcapacity = infra.recieve_and_convert_log.computecapacity(log)
initiallimit = infra.recieve_and_convert_log.initiallimit(log)[0]
initialcaplim = []
for i in range(len(initialcapacity)):
initialcaplim.append(
(initialcapacity[i][0], initialcapacity[i][1], initiallimit[i][1]))
#print(intialcapacity,"line 205")
Actresource = roles_discovery.apply(
log, variant=None, parameters={rpd.Parameters.RESOURCE_KEY: logreso})
list0 = []
infra.recieve_and_convert_log.notdoact(ptree, list0)
handover = infra.recieve_and_convert_log.getactivityresourcecount(
log, list0, logname, logreso)[1]
for i, x in enumerate(duration):
duration[i] = (x[0], round(x[1], 2), round(deviation[i][1], 2))
for i, x in enumerate(deviation):
deviation[i] = (x[0], round(x[1], 2))
context = {'log':log,'ptree':ptree,'duration':duration,'deviation':deviation,\
'worked_time':worked_time,'numtrace':numtrace,'numactivity':numactivity,'activitylist':activitylist,\
'timeinterval':timeinterval,'meanthoughputtime':meanthoughputtime,\
'deviationthoughputtime':deviationthoughputtime,'arrivalratio':arrivalratio,\
'dispersionratio':dispersionratio,'resourcedict':Actresource,'handover':handover,"initialcaplim":initialcaplim,'initialtrace':initialtrace}
return render(request, 'statics.html', context)
def performance_calculation_ocel_aggregation(ocel: OCEL, aggregation: Dict[str, Dict[Tuple[str, str], Set[Any]]],
parameters: Optional[Dict[Any, Any]] = None) -> Dict[
str, Dict[Tuple[str, str], List[float]]]:
"""
Calculates the performance based on one of the following aggregations:
- aggregate_ev_couples
- aggregate_total_objects
Parameters
----------------
ocel
Object-centric event log
aggregation
Aggregation calculated using one of the aforementioned methods
parameters
Parameters of the algorithm, including:
- Parameters.EVENT_ID => the event identifier
- Parameters.EVENT_TIMESTAMP => the timestamp
- Parameters.BUSINESS_HOURS => enables/disables the business hours
- Parameters.WORKTIMING => the work timing (default: [7, 17])
- Parameters.WEEKENDS => the weekends (default: [6, 7])
Returns
----------------
edges_performance
For each object type, associate a dictionary where to each activity couple
all the times between the activities are recorded.
"""
if parameters is None:
parameters = {}
event_id = exec_utils.get_param_value(Parameters.EVENT_ID, parameters, ocel.event_id_column)
timestamp_key = exec_utils.get_param_value(Parameters.EVENT_TIMESTAMP, parameters, ocel.event_timestamp)
timestamps = ocel.events.groupby(event_id)[timestamp_key].apply(list).to_dict()
timestamps = {x: y[0] for x, y in timestamps.items()}
business_hours = exec_utils.get_param_value(Parameters.BUSINESS_HOURS, parameters, False)
worktiming = exec_utils.get_param_value(Parameters.WORKTIMING, parameters, [7, 17])
weekends = exec_utils.get_param_value(Parameters.WEEKENDS, parameters, [6, 7])
workcalendar = exec_utils.get_param_value(Parameters.WORKCALENDAR, parameters, constants.DEFAULT_BUSINESS_HOURS_WORKCALENDAR)
ret = {}
for ot in aggregation:
ret[ot] = {}
for act in aggregation[ot]:
ret[ot][act] = []
for el in aggregation[ot][act]:
if business_hours:
bh = BusinessHours(timestamps[el[0]],
timestamps[el[1]],
worktiming=worktiming,
weekends=weekends,
workcalendar=workcalendar)
diff = bh.getseconds()
else:
diff = timestamps[el[1]].timestamp() - timestamps[el[0]].timestamp()
ret[ot][act].append(diff)
ret[ot][act] = sorted(ret[ot][act])
return ret
def single_element_statistics(log,
net,
initial_marking,
aligned_traces,
variants_idx,
activity_key="concept:name",
timestamp_key="time:timestamp",
ht_perf_method="last",
parameters=None):
"""
Get single Petrinet element statistics
Parameters
------------
log
Log
net
Petri net
initial_marking
Initial marking
aligned_traces
Result of the token-based replay
variants_idx
Variants along with indexes of belonging traces
activity_key
Activity key (must be specified if different from concept:name)
timestamp_key
Timestamp key (must be specified if different from time:timestamp)
ht_perf_method
Method to use in order to annotate hidden transitions (performance value could be put on the last possible
point (last) or in the first possible point (first)
parameters
Possible parameters of the algorithm
Returns
------------
statistics
Petri net element statistics (frequency, unaggregated performance)
"""
if parameters is None:
parameters = {}
business_hours = parameters[
"business_hours"] if "business_hours" in parameters else False
worktiming = parameters["worktiming"] if "worktiming" in parameters else [
7, 17
]
weekends = parameters["weekends"] if "weekends" in parameters else [6, 7]
statistics = {}
for variant in variants_idx:
first_trace = log[variants_idx[variant][0]]
act_trans = aligned_traces[variants_idx[variant]
[0]]["activated_transitions"]
annotations_places_trans, annotations_arcs = calculate_annotation_for_trace(
first_trace,
net,
initial_marking,
act_trans,
activity_key,
ht_perf_method=ht_perf_method)
for el in annotations_places_trans:
if el not in statistics:
statistics[el] = {
"count": 0,
"performance": [],
"log_idx": [],
"no_of_times_enabled": 0,
"no_of_times_activated": 0
}
statistics[el]["count"] += annotations_places_trans[el][
"count"] * len(variants_idx[variant])
if "no_of_times_enabled" in annotations_places_trans[el]:
statistics[el][
"no_of_times_enabled"] += annotations_places_trans[el][
"no_of_times_enabled"] * len(variants_idx[variant])
statistics[el][
"no_of_times_activated"] += annotations_places_trans[el][
"no_of_times_activated"] * len(variants_idx[variant])
if "performance" in annotations_places_trans[el]:
for trace_idx in variants_idx[variant]:
trace = log[trace_idx]
for perf_couple in annotations_places_trans[el][
"performance"]:
if timestamp_key in trace[perf_couple[
0]] and timestamp_key in trace[perf_couple[1]]:
if business_hours:
bh = BusinessHours(
trace[perf_couple[1]]
[timestamp_key].replace(tzinfo=None),
trace[perf_couple[0]]
[timestamp_key].replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
perf = bh.getseconds()
else:
perf = (trace[perf_couple[0]][timestamp_key] -
trace[perf_couple[1]][timestamp_key]
).total_seconds()
else:
perf = 0.0
statistics[el]["performance"].append(perf)
statistics[el]["log_idx"].append(trace_idx)
for el in annotations_arcs:
if el not in statistics:
statistics[el] = {"count": 0, "performance": []}
statistics[el]["count"] += annotations_arcs[el]["count"] * len(
variants_idx[variant])
for trace_idx in variants_idx[variant]:
trace = log[trace_idx]
for perf_couple in annotations_arcs[el]["performance"]:
if timestamp_key in trace[perf_couple[
0]] and timestamp_key in trace[perf_couple[1]]:
if business_hours:
bh = BusinessHours(
trace[perf_couple[1]][timestamp_key].replace(
tzinfo=None),
trace[perf_couple[0]][timestamp_key].replace(
tzinfo=None),
worktiming=worktiming,
weekends=weekends)
perf = bh.getseconds()
else:
perf = (trace[perf_couple[0]][timestamp_key] -
trace[perf_couple[1]][timestamp_key]
).total_seconds()
else:
perf = 0.0
statistics[el]["performance"].append(perf)
return statistics
def apply(
log: EventLog,
temporal_profile: typing.TemporalProfile,
parameters: Optional[Dict[Any, Any]] = None
) -> typing.TemporalProfileConformanceResults:
"""
Checks the conformance of the log using the provided temporal profile.
Implements the approach described in:
Stertz, Florian, Jürgen Mangler, and Stefanie Rinderle-Ma. "Temporal Conformance Checking at Runtime based on Time-infused Process Models." arXiv preprint arXiv:2008.07262 (2020).
Parameters
---------------
log
Event log
temporal_profile
Temporal profile
parameters
Parameters of the algorithm, including:
- Parameters.ACTIVITY_KEY => the attribute to use as activity
- Parameters.START_TIMESTAMP_KEY => the attribute to use as start timestamp
- Parameters.TIMESTAMP_KEY => the attribute to use as timestamp
- Parameters.ZETA => multiplier for the standard deviation
- Parameters.BUSINESS_HOURS => calculates the difference of time based on the business hours, not the total time.
Default: False
- Parameters.WORKTIMING => work schedule of the company (provided as a list where the first number is the start
of the work time, and the second number is the end of the work time), if business hours are enabled
Default: [7, 17] (work shift from 07:00 to 17:00)
- Parameters.WEEKENDS => indexes of the days of the week that are weekend
Default: [6, 7] (weekends are Saturday and Sunday)
Returns
---------------
list_dev
A list containing, for each trace, all the deviations.
Each deviation is a tuple with four elements:
- 1) The source activity of the recorded deviation
- 2) The target activity of the recorded deviation
- 3) The time passed between the occurrence of the source activity and the target activity
- 4) The value of (time passed - mean)/std for this occurrence (zeta).
"""
if parameters is None:
parameters = {}
log = log_converter.apply(log,
variant=log_converter.Variants.TO_EVENT_LOG,
parameters=parameters)
business_hours = exec_utils.get_param_value(Parameters.BUSINESS_HOURS,
parameters, False)
worktiming = exec_utils.get_param_value(Parameters.WORKTIMING, parameters,
[7, 17])
weekends = exec_utils.get_param_value(Parameters.WEEKENDS, parameters,
[6, 7])
workcalendar = exec_utils.get_param_value(
Parameters.WORKCALENDAR, parameters,
constants.DEFAULT_BUSINESS_HOURS_WORKCALENDAR)
activity_key = exec_utils.get_param_value(Parameters.ACTIVITY_KEY,
parameters,
xes_constants.DEFAULT_NAME_KEY)
timestamp_key = exec_utils.get_param_value(
Parameters.TIMESTAMP_KEY, parameters,
xes_constants.DEFAULT_TIMESTAMP_KEY)
start_timestamp_key = exec_utils.get_param_value(
Parameters.START_TIMESTAMP_KEY, parameters,
xes_constants.DEFAULT_TIMESTAMP_KEY)
zeta = exec_utils.get_param_value(Parameters.ZETA, parameters, 6.0)
ret = []
for trace in log:
deviations = []
for i in range(len(trace) - 1):
act_i = trace[i][activity_key]
time_i = trace[i][timestamp_key].timestamp()
for j in range(i + 1, len(trace)):
time_j = trace[j][start_timestamp_key].timestamp()
if time_j >= time_i:
act_j = trace[j][activity_key]
if (act_i, act_j) in temporal_profile:
if business_hours:
bh = BusinessHours(
trace[i][timestamp_key].replace(tzinfo=None),
trace[j][start_timestamp_key].replace(
tzinfo=None),
worktiming=worktiming,
weekends=weekends,
workcalendar=workcalendar)
this_diff = bh.getseconds()
else:
this_diff = time_j - time_i
mean = temporal_profile[(act_i, act_j)][0]
std = temporal_profile[(act_i, act_j)][1]
if this_diff < mean - zeta * std or this_diff > mean + zeta * std:
this_zeta = abs(this_diff - mean
) / std if std > 0 else sys.maxsize
deviations.append(
(act_i, act_j, this_diff, this_zeta))
ret.append(deviations)
return ret
def get_cases_description(log, parameters=None):
"""
Get a description of traces present in the log_skeleton
Parameters
-----------
log
Log
parameters
Parameters of the algorithm, including:
Parameters.CASE_ID_KEY -> Trace attribute in which the case ID is contained
Parameters.TIMESTAMP_KEY -> Column that identifies the timestamp
Parameters.ENABLE_SORT -> Enable sorting of traces
Parameters.SORT_BY_INDEX -> Sort the traces using this index:
0 -> case ID
1 -> start time
2 -> end time
3 -> difference
Parameters.SORT_ASCENDING -> Set sort direction (boolean; it true then the sort direction is ascending, otherwise
descending)
Parameters.MAX_RET_CASES -> Set the maximum number of returned traces
Returns
-----------
ret
Dictionary of traces associated to their start timestamp, their end timestamp and their duration
"""
if parameters is None:
parameters = {}
case_id_key = exec_utils.get_param_value(Parameters.CASE_ID_KEY, parameters, DEFAULT_TRACEID_KEY)
timestamp_key = exec_utils.get_param_value(Parameters.TIMESTAMP_KEY, parameters, DEFAULT_TIMESTAMP_KEY)
enable_sort = exec_utils.get_param_value(Parameters.ENABLE_SORT, parameters, True)
sort_by_index = exec_utils.get_param_value(Parameters.SORT_BY_INDEX, parameters, 0)
sort_ascending = exec_utils.get_param_value(Parameters.SORT_ASCENDING, parameters, True)
max_ret_cases = exec_utils.get_param_value(Parameters.MAX_RET_CASES, parameters, None)
business_hours = exec_utils.get_param_value(Parameters.BUSINESS_HOURS, parameters, False)
worktiming = exec_utils.get_param_value(Parameters.WORKTIMING, parameters, [7, 17])
weekends = exec_utils.get_param_value(Parameters.WEEKENDS, parameters, [6, 7])
statistics_list = []
for index, trace in enumerate(log):
if trace:
ci = trace.attributes[case_id_key] if case_id_key in trace.attributes else "EMPTY" + str(index)
st = trace[0][timestamp_key]
et = trace[-1][timestamp_key]
if business_hours:
bh = BusinessHours(st.replace(tzinfo=None), et.replace(tzinfo=None), worktiming=worktiming,
weekends=weekends)
diff = bh.getseconds()
else:
diff = et.timestamp() - st.timestamp()
st = st.timestamp()
et = et.timestamp()
statistics_list.append([ci, st, et, diff])
if enable_sort:
statistics_list = sorted(statistics_list, key=lambda x: x[sort_by_index], reverse=not sort_ascending)
if max_ret_cases is not None:
statistics_list = statistics_list[:min(len(statistics_list), max_ret_cases)]
statistics_dict = {}
for el in statistics_list:
statistics_dict[str(el[0])] = {"startTime": el[1], "endTime": el[2], "caseDuration": el[3]}
return statistics_dict
def get_cases_description(log, parameters=None):
"""
Get a description of traces present in the log
Parameters
-----------
log
Log
parameters
Parameters of the algorithm, including:
case_id_key -> Trace attribute in which the case ID is contained
timestamp_key -> Column that identifies the timestamp
enable_sort -> Enable sorting of traces
sort_by_index -> Sort the traces using this index:
0 -> case ID
1 -> start time
2 -> end time
3 -> difference
sort_ascending -> Set sort direction (boolean; it true then the sort direction is ascending, otherwise
descending)
max_ret_cases -> Set the maximum number of returned traces
Returns
-----------
ret
Dictionary of traces associated to their start timestamp, their end timestamp and their duration
"""
if parameters is None:
parameters = {}
case_id_key = parameters[
PARAMETER_CONSTANT_CASEID_KEY] if PARAMETER_CONSTANT_CASEID_KEY in parameters else DEFAULT_TRACEID_KEY
timestamp_key = parameters[
PARAMETER_CONSTANT_TIMESTAMP_KEY] if PARAMETER_CONSTANT_TIMESTAMP_KEY in parameters else DEFAULT_TIMESTAMP_KEY
enable_sort = parameters[
"enable_sort"] if "enable_sort" in parameters else True
sort_by_index = parameters[
"sort_by_index"] if "sort_by_index" in parameters else 0
sort_ascending = parameters[
"sort_ascending"] if "sort_ascending" in parameters else True
max_ret_cases = parameters[
"max_ret_cases"] if "max_ret_cases" in parameters else None
business_hours = parameters[
"business_hours"] if "business_hours" in parameters else False
worktiming = parameters["worktiming"] if "worktiming" in parameters else [
7, 17
]
weekends = parameters["weekends"] if "weekends" in parameters else [6, 7]
statistics_list = []
for index, trace in enumerate(log):
if trace:
ci = trace.attributes[
case_id_key] if case_id_key in trace.attributes else "EMPTY" + str(
index)
st = trace[0][timestamp_key]
et = trace[-1][timestamp_key]
if business_hours:
bh = BusinessHours(st.replace(tzinfo=None),
et.replace(tzinfo=None),
worktiming=worktiming,
weekends=weekends)
diff = bh.getseconds()
else:
diff = et.timestamp() - st.timestamp()
st = st.timestamp()
et = et.timestamp()
statistics_list.append([ci, st, et, diff])
if enable_sort:
statistics_list = sorted(statistics_list,
key=lambda x: x[sort_by_index],
reverse=not sort_ascending)
if max_ret_cases is not None:
statistics_list = statistics_list[:min(len(statistics_list
), max_ret_cases)]
statistics_dict = {}
for el in statistics_list:
statistics_dict[str(el[0])] = {
"startTime": el[1],
"endTime": el[2],
"caseDuration": el[3]
}
return statistics_dict
