The swaying “having the right person at the right time at the right place” makes even more sense in a low-oil-price environment. Understanding capabilities available in an organization can be hindered by the size of the talent pool and a lack of a structured and systematic approach to quantify, update, and manage competencies.
This paper describes a novel method based on machine learning to maintain an evergreen competency database. The tool reduces discrepancies between organizational requirements and the actual talent deployment by using unstructured corporate data.
A digital database for each person was created and populated with various data such as CV, corporate information (e.g., compensation and location), assignment history, technical reports generated, articles and conference papers written, training attended, and periodic internal competency assessments and promotion history. A master database then is created that includes the database of each person. Machine-learning methods based on optical characters recognition and natural-language processing then are applied to the master database to extract synthetic and key information needed for more-efficient resource management and capability deployment.
For each employee, a competency matrix is created following a multitiered proficiency system with levels ranging from awareness to mastery. The machine-learning methods result in a cloud of keywords that can be browsed. Managers are now capable of having a snapshot of their current workforce skills and capabilities according to competency level.
External workforce such as former employees or prospective candidates can be included in the capability pool. In addition, a focused search, based on specific project needs and niche capability requirements, can be conducted.
In a few clicks, a talent manager can look up critical resources that can be deployed when and where needed. For entry-level resources, a personal development plan can be drafted with more granularity and can be tailored by the capabilities available at a given location or operating unit at a given time.