def apply(mco, parameters=None):
"""
Calculate the Handover of Work metric
Parameters
------------
mco
Matrix container object
parameters
Parameters of the algorithm
Returns
------------
rsc_rsc_matrix
Resource-Resource Matrix containing the Real Handover of Work metric value
"""
if parameters is None:
parameters = {}
rsc_rsc_matrix = rscrsc_utils.get_empty_rscrsc_matrix(mco)
for resource, next_resource in zip(mco.dataframe['resource'],
mco.dataframe['next_resource']):
rsc_rsc_matrix[mco.resources_list.index(resource)][
mco.resources_list.index(next_resource)] += 1
return rsc_rsc_matrix
def apply(mco, parameters=None):
"""
Calculate the Similar Activities metric
Parameters
------------
mco
Matrix container object
parameters
Parameters of the algorithm
Returns
------------
rsc_rsc_matrix
Resource-Resource Matrix containing the Similar Activities metric value
"""
if parameters is None:
parameters = {}
if len(mco.activities_list) == 0:
raise Exception("must provide full MCO dataframe")
all_resources = np.concatenate(
[mco.dataframe['resource'], mco.dataframe['next_resource']], axis=0)
all_activities = np.concatenate(
[mco.dataframe['activity'], mco.dataframe['next_activity']], axis=0)
all_resource_activity = np.concatenate([
all_resources.reshape(len(all_resources), 1),
all_activities.reshape(len(all_activities), 1)
],
axis=1)
all_resource_activity_df = pd.DataFrame(columns=['resource', 'activity'],
data=all_resource_activity)
grouped_by_activity = all_resource_activity_df.groupby(['resource'])
resource_activity_dict = defaultdict(list)
for resource, group in grouped_by_activity:
resource_activity_dict[resource].append(group['activity'].values)
rsc_act_matrix = rscact_utils.get_empty_rscact_matrix(mco)
for resource in resource_activity_dict:
for activity_list in resource_activity_dict[resource]:
for activity in activity_list:
rsc_act_matrix[mco.resources_list.index(resource)][
mco.activities_list.index(activity)] += 1
rsc_rsc_matrix = rscrsc_utils.get_empty_rscrsc_matrix(mco)
for index, resource in enumerate(rsc_act_matrix):
for rest in range(index + 1, rsc_act_matrix.shape[0]):
main_resource = resource
other_resource = rsc_act_matrix[rest]
r, p = pearsonr(main_resource, other_resource)
rsc_rsc_matrix[index][rest] = r
return rsc_rsc_matrix
def apply(mco, parameters=None):
"""
Calculate the Real Handover of Work metric
Parameters
------------
mco
Matrix container object
parameters
Parameters of the algorithm
Returns
------------
rsc_rsc_matrix
Resource-Resource Matrix containing the Real Handover of Work metric value
"""
if parameters is None:
parameters = {}
dependency_threshold = parameters[
"dependency_threshold"] if "dependency_threshold" in parameters else 0
if len(mco.activities_list) == 0:
raise Exception("must provide full MCO dataframe")
rsc_rsc_matrix = rscrsc_utils.get_empty_rscrsc_matrix(mco)
grouped_activity_paris = mco.dataframe.groupby(
['activity', 'next_activity']).size().reset_index(name='counts')
for index, row in mco.dataframe.iterrows():
ab = grouped_activity_paris.loc[
(grouped_activity_paris['activity'] == row['activity']) &
(grouped_activity_paris['next_activity'] == row['next_activity'])]
if ab.empty:
abn = 0
else:
abn = ab.iloc[0, 2]
ba = grouped_activity_paris.loc[
(grouped_activity_paris['next_activity'] == row['activity'])
& (grouped_activity_paris['activity'] == row['next_activity'])]
if ba.empty:
ban = 0
else:
ban = ab.iloc[0, 2]
if row['activity'] == row['next_activity']:
dependency = abn / (abn + 1)
else:
dependency = (abn - ban) / (abn + ban + 1)
if dependency > dependency_threshold:
rsc_rsc_matrix[mco.resources_list.index(
row['resource'])][mco.resources_list.index(
row['next_resource'])] += 1
return rsc_rsc_matrix