| 主权项 |
1. A method of prioritizing work requests to resolve incidents in an information technology (IT) infrastructure, the method comprising the steps of:
a computer dividing, into first and second data sets, historical data, which includes historical work requests that requested resolutions of historical incidents in the IT infrastructure, each of the work requests comprising data fields; the computer generating, using a first set of data fields of the work requests in the first data set, one or more incident concepts; the computer combining the one or more incident concepts with a second set of data fields of the work requests in the first data set to form a set of predictive variables; based on the predictive variables, the computer generating a model which associates the historical data with a resolution of a historical incident which caused a problem in the IT infrastructure and determining the resolution of the historical incident occurred in a time period, the historical data including a first keyword included in a description of a historical work request that requested a resolution of the historical incident which caused a problem in the IT infrastructure and at least one of: (1) a first set of skills required by a person to resolve the historical incident, (2) a first customer who initiated the historical work request, (3) a first classification indicating a component of the IT infrastructure affected by the historical incident, (4) a first arrival date and time at which the historical work request was initiated by the customer, (5) a first level of severity of the historical incident based on a number of users affected by the historical incident or a potential amount of business lost as a result of the historical incident, (6) a first complexity of the historical incident, and (7) a first target completion date and time at which the historical incident was expected to be resolved; the computer validating the model using the second data set; the computer determining data elements included in a current work request that requests a resolution of a current incident affecting the IT infrastructure, the data elements including a second keyword included in a description of the current work request and at least one of: (1) a second set of skills required by a person to resolve the current incident, (2) a second customer who initiated the current work request, (3) a second classification indicating a component of the IT infrastructure affected by the current incident, (4) a second arrival date and time at which the current work request was initiated by the customer, (5) a second level of severity of the current incident based on a number of users affected by the current incident or a potential amount of business lost as a result of the current incident, and (6) a second complexity of the current incident; the computer determining a match between the first and second keywords, and determining at least one match included in a match between the first and second sets of skills, a match between the first and second customers, a match between the first and second classifications, a match between a day of a week specified by the first arrival date and a day of the week specified by the second arrival date, a match between the first and second levels of severity, and a match between the first and second complexities; based on the model, the match between the first and second keywords, and the at least one match included in the matches between the first and second sets of skills, the first and second customers, the first and second classifications, the day of the week specified by the first arrival date and the day of the week specified by the second arrival date, the first and second levels of severity, and the first and second complexities, the computer predicting a resolution of the current incident will occur within the time period, which exceeds a maximum amount of time permitted to resolve the current incident, the maximum amount of time being specified in a service level agreement with the second customer; and based at least in part on the current incident predicted to be resolved within the time period which exceeds the maximum amount of time permitted to resolve the current incident, the computer assigning a priority to the current work request that is higher than a priority assigned to another current work request that requests a resolution of another current incident in the IT infrastructure which, based at least in part on the model, is predicted to occur within another time period that does not exceed another maximum amount of time permitted to resolve the other current incident, the other maximum amount of time being specified in another service level agreement. |
