Abstract: A method is provided, including: storing comments generated in response to a content item served over a network; analyzing the comments to determine features associated with each of the comments; using a scoring model to score each comment based on the comment's corresponding features; receiving a request to serve a subset of the comments; responsive to the request, selecting a ranking of the comments that is one permutation from possible rankings of the comments, wherein selecting the ranking is in accordance with a probability distribution of the possible rankings that is based on the scores of the comments; serving comments identified by the selected ranking over the network to a client device; determining a dwell time on the served comments; applying the dwell time to update the scoring model.

Abstract: A method is provided, including: storing comments generated in response to a content item served over a network; analyzing the comments to determine features associated with each of the comments; using a scoring model to score each comment based on the comment's corresponding features; receiving a request to serve a subset of the comments; responsive to the request, selecting a ranking of the comments that is one permutation from possible rankings of the comments, wherein selecting the ranking is in accordance with a probability distribution of the possible rankings that is based on the scores of the comments; serving comments identified by the selected ranking over the network to a client device; determining a dwell time on the served comments; applying the dwell time to update the scoring model.

Abstract: A component testing plan considers both distinguishable components and undistinguishable components for each of the distinguishable components. In addition to distinguishable and undistinguishable components, inputs to the system include a level of customer demand, using component-based demand forecasting, and what test types are to be performed. The system then determines a set of combinatorial test options for each of the test types and a cycle time for each combinatorial test option. The system then cognitively develops a component testing plan including a listing of combinatorial test(s) of the set of combinatorial test options for each test type to be performed and how many times to perform each combinatorial test(s) on a testing machine with a known capacity in order to minimize a total cycle time for all of the test types and a number of the testing machine needed; and implementing the component testing plan on a set of test components on a production line.

Abstract: A method is provided, including: storing comments generated in response to a content item served over a network; analyzing the comments to determine features associated with each of the comments; using a scoring model to score each comment based on the comment's corresponding features; receiving a request to serve a subset of the comments; responsive to the request, selecting a ranking of the comments that is one permutation from possible rankings of the comments, wherein selecting the ranking is in accordance with a probability distribution of the possible rankings that is based on the scores of the comments; serving comments identified by the selected ranking over the network to a client device; determining a dwell time on the served comments; applying the dwell time to update the scoring model.

Abstract: Disclosed is a computer-implemented method of employee/resource allocation across an enterprise. The method includes receiving, by a data processing system, a workload matrix, the workload matrix including job types and a level of demand for each job type; receiving, by the data processing system, an employee matrix, wherein the employee matrix comprises a plurality of employees and a plurality of characteristics associated with each of the plurality of employees; generating, by the data processing system, a skills matrix for the enterprise based on the plurality of characteristics associated with each of the plurality of employees; cognitively generating, by the data processing system, an optimization model; responsive to performing an iteration of the optimization model, assigning, by the data processing system, the plurality of employees to the plurality of job types based on the iteration, resulting in a report; and implementing the report in the enterprise.

Abstract: A component testing plan considers both distinguishable components and undistinguishable components for each of the distinguishable components. In addition to distinguishable and undistinguishable components, inputs to the system include a level of customer demand, using component-based demand forecasting, and what test types are to be performed. The system then determines a set of combinatorial test options for each of the test types and a cycle time for each combinatorial test option. The system then cognitively develops a component testing plan including a listing of combinatorial test(s) of the set of combinatorial test options for each test type to be performed and how many times to perform each combinatorial test(s) on a testing machine with a known capacity in order to minimize a total cycle time for all of the test types and a number of the testing machine needed; and implementing the component testing plan on a set of test components on a production line.

Abstract: Technical solutions are described to for assigning a team to resolve a defect during a manufacturing process. The method includes identifying a root cause of a defect from a root cause database by generating a reference code. The method also includes generating a defect score based on characteristics of the defect and the root cause, and identifying a set of solutions for the defect based on the defect score. The method also includes selecting a first solution from the set of solutions to implement and determining a candidate skill for the implementation. The method also includes identifying, from an employee database, a candidate employee that has the candidate skill and assigning the candidate employee to the team to resolve the defect.

Abstract: A method, data processing system and computer program product for managing inventory used in a manufacturing operation to manufacture a product. The method includes a processor communicating with at least one database containing manufacturing operation data. Several inventory locations and components are identified within the manufacturing operation data. A current component inventory and space data, historical component usage data and a demand forecast are retrieved. A statistical analysis is performed upon the current component inventory and space data, the historical component usage data and the demand forecast to obtain a target inventory level range for each of the components at each of the inventory locations. An inventory level control chart is generated, based on the statistical analysis, for selected components associated with at least one inventory location.

Abstract: Technical solutions are described to for assigning a team to resolve a defect during a manufacturing process. The method includes identifying a root cause of a defect from a root cause database by generating a reference code. The method also includes generating a defect score based on characteristics of the defect and the root cause, and identifying a set of solutions for the defect based on the defect score. The method also includes selecting a first solution from the set of solutions to implement and determining a candidate skill for the implementation. The method also includes identifying, from an employee database, a candidate employee that has the candidate skill and assigning the candidate employee to the team to resolve the defect.

Abstract: A set of channel inputs corresponding to an order is received. A subset of the set is transformed, using a transformation rule, into a set of strategic transformed inputs. A total impact value and an integrated risk likelihood value corresponding to the order are computed. The integrated risk likelihood value and the total impact value of the order are output, to a shipping system, relative to a second integrated risk likelihood value and a second total impact value of a second completed order from a second channel, as a comparison of a first risk associated with the completed order relative to a second risk associated with the second completed order at a shipping dock. The shipping dock is shared by the order processing channel and the second order processing channel. When the first risk being higher than the second risk, the order is selected for shipping using the shipping system.

Abstract: Embodiments of the invention relate to inventory and resource sharing for sustaining a build-to-order supply chain. A multi-objective optimization model is formulated to identify an optimal strategy that is adaptive and responsive to customer demand to maintain and meet orders in the supply chain while mitigating demand and supply risks.

Abstract: Embodiments of the invention relate to an attribute based approach for assessment ergonomic risks, and applying the assessment to reduce the identified ergonomic risks. Ergonomic risks scores are calculated for risk type, operator, and workplace areas. Risks levels are minimized through mitigation strategies and effectiveness and ease of implementation. An operator is assigned or re-allocated to a designated workplace area based on consideration of both workload and ergonomic risk.

Abstract: A method, data processing system and computer program product for managing inventory used in a manufacturing operation to manufacture a product. The method includes a processor communicating with at least one database containing manufacturing operation data. Several inventory locations and components are identified within the manufacturing operation data. A current component inventory and space data, historical component usage data and a demand forecast are retrieved. A statistical analysis is performed upon the current component inventory and space data, the historical component usage data and the demand forecast to obtain a target inventory level range for each of the components at each of the inventory locations. An inventory level control chart is generated, based on the statistical analysis, for selected components associated with at least one inventory location.

Abstract: A method for enabling enhanced decision-making when shipping parts between sites within an organization includes receiving a plurality of orders to deliver parts from a first site to a second site. The method determines a shipping option for shipping the parts from the first site to the second site and, for each of the orders, a transportation risk associated with the shipping option. The transportation risk varies in accordance with a probability that the shipping option will result in a delay, and an amount of revenue that will be affected as a result of the delay. The transportation risk for each of the orders is displayed in a matrix. The method further enables a user to modify the shipping option to adjust the position of each transportation risk within the matrix. A corresponding apparatus and computer program product are also disclosed.

Abstract: A method and system are disclosed for determining the locations of parts in a warehouse. The method comprises the step of generating for each part an associated designator having first and second components, including the steps of, for each of the parts, (i) determining the first component of the associated designator based on physical dimensions of the part, and (ii) determining the second component of the associated designator based on a pick velocity of the part. These designators are used to identify locations in the warehouse for the parts. In a preferred embodiment, the first components of the designators are used to determine the distances of the identified locations for the parts from an entrance/exit of the warehouse, and the second components of the designators are used to determine the heights of the identified locations for the parts in the warehouse.

Abstract: A total cost estimate is calculated based on a set of printed circuit board (PCB) design parameters. The set of PCB design parameters are received, and PCB attributes are extracted from them. Based on the PCB attributes the PCB is classified and a cost equation is calculated. The cost equation is calculated based on a regression analysis of one or more of the PCB attributes. Once the cost equation is calculated, the total cost is computed based on the cost equation.

Abstract: An embodiment of the invention provides a method for storing a cable, including receiving one or more cables on an adjustable central coil. An RFID chip connected to the cable is read via a RFID reader proximate the central coil. The cable is routed from the central coil to one or more adjustable middle coils, and routed from the middle coil to an adjustable outer coil. The cable is dispensed from the outer coil; and, the RFID chip is read via a RFID reader proximate the outer coil. The method modifies the distance between the central coil and the middle coil, the central coil and the outer coil, and/or the middle coil and the outer coil. Specifically, the diameter of the central coil, the middle coil, and/or the outer coil is adjusted. The modification accommodates for the arc limit and/or head size of the cable.

Abstract: A computer-automated method for analyzing an organizational floorplan layout, and making recommendations for modifying the layout to optimize productivity, and efficiency of operations conducted within the modified layout includes the following method steps. Gathering raw performance measure date comprising performance measures known to quantify the floorplan layout productivity and efficiency of operations, performing statistical analyses on the raw performance measure data and comparing different areas in the floorplan layout using results of the statistical analyses, and generating proximity data as a result of the comparing.

Abstract: An embodiment of the invention provides a method for storing a cable, including receiving one or more cables on an adjustable central coil. An RFID chip connected to the cable is read via a RFID reader proximate the central coil. The cable is routed from the central coil to one or more adjustable middle coils, and routed from the middle coil to an adjustable outer coil. The cable is dispensed from the outer coil; and, the RFID chip is read via a RFID reader proximate the outer coil. The method modifies the distance between the central coil and the middle coil, the central coil and the outer coil, and/or the middle coil and the outer coil. Specifically, the diameter of the central coil, the middle coil, and/or the outer coil is adjusted. The modification accommodates for the arc limit and/or head size of the cable.

Abstract: A method for effective employee assignment including determining a number of employees needed to perform each of a plurality of tasks, determining a skill flexibility index (SFI) of each employee, determining a job quality index (JQI) of each employee, determining a minimum acceptable (min. JQI) for each task, and for each task, in an order from highest min. JQI to lowest min. JQI, assigning each employee to the task having a JQI higher than or equivalent to the min. JQI in an order from each employee with a lower SFI and a higher JQI to an employee with a higher SFI and a lower JQI until the number of employees needed to perform the task is assigned or there are no more available employees to assign to the task having a JQI higher than or equivalent to the min. JQI of the task. A corresponding system and computer program product.