Timing estimations for application lifecycle management work items determined through machine learning

A non-transitory machine-readable medium comprising instructions executable by a processing resource to:

• access data records of completed work items associated with managing a lifecycle of a software application;
• extract feature values from the data records of the completed work items for a selected set of features;
• determine timing data of state transitions for the completed work items from the data records;
• use machine learning with the timing data and the extracted feature values as input samples to generate predictors of transition probabilities of state transitions for work items with various feature values for the extracted feature values;
• use the predictors to create a transition matrix for a work item with specific feature values, wherein the transition matrix specifies transition probabilities for each state transition that the work item traverses and includes rows and columns,
• wherein the transition matrix includes a first value at a first row and a first column and a second value at the first row and a second column, the first value corresponding to a length of time that the work item remained in a first state and the second value corresponding to a length of time the work item remained in the first state prior to transitioning from the first state to the second state;
• and
• reference the transition matrix to determine an estimated timing of a particular state transition for the uncompleted work item.

A system comprising:

• a processor;
• and
• a memory on which is stored machine readable instructions that are to cause the processor to:
• access data records of completed work items associated with managing a lifecycle of a software application;
• extract feature values from the data records of the completed work items for a selected set of features;
• determine timing data of state transitions for the completed work items from the data records;
• access a transition matrix for a work item with specific feature values, wherein the transition matrix specifies transition probabilities for each state transition that the work item with the specific feature values traverses and includes rows and columns,
• wherein the transition matrix includes a first value at a first row and a first column and a second value at the first row and a second column, the first value corresponding to a length of time that the work item remained in a first state and the second value corresponding to one of a number of times that the work item transitioned from the first state to a second state or a length of time that the work item remained in the first state prior to transitioning from the first state to the second state, and wherein the transition probabilities are estimated through machine learning using the timing data and the extracted feature values as input samples;
• and
• reference the transition matrix to determine an estimated timing of state transitions for an uncompleted work item with the specific feature values.