Hybrid cloud migration delay risk prediction engine

A method comprising:

• storing activity data for a cloud migration project in a wave report database;
• storing historical activity data for previously completed cloud migration projects in an all activities database;
• grouping data assigned for migration in bundle units according to bundling rules, the bundling rules comprising client input parameters and historical bundling rules;
• assigning the bundle units to a cloud migration wave according to wave assignment rules, the wave assignment rules comprising client input parameters and historical wave assignment rules;
• sequencing, by an analytics engine, activities included in the activity data according to a respective activity deadline;
• determining, by the analytics engine, buffer times between subsequent activities in the activity data;
• determining, by the analytics engine, delays for activities in the activity data;
• determining, by the analytics engine, a view value for activities in the activity data based on a corresponding time to data wave migration;
• determining, by the analytics engine, an activity value for activities in the activity data based on a corresponding activity status;
• determining, by the analytics engine, a predicted delay of the cloud migration project based on at least a first iteration and a second iteration of a random forest model on the historical activity data for previously completed cloud migration projects, wherein the first iteration selects an optimum number of data trees in the random forest model, and the second iteration fine tunes model arguments in the random forest model;
• and
• determining, by the analytics engine, a predicted delay score based on a weighted consideration of the predicted delay, the view value, and the activity value.

A product comprising:

• a machine-readable medium other than a transitory signal;
• and
• instructions stored on the machine-readable medium, the instructions configured to, when executed, cause a machine to:
• at network interface circuitry:
• receive activity data for a cloud migration project;
• and
• receive historical activity data for previously completed cloud migration projects;
• and
• at an analytics engine:
• group data assigned for migration in bundle units according to bundling rules, the bundling rules comprising client input parameters and historical bundling rules;
• assign the bundle units to a cloud migration wave according to wave assignment rules, the wave assignment rules comprising client input parameters and historical wave assignment rules;
• sequence activities included in the activity data according to a respective activity deadline;
• determine buffer times between subsequent activities in the activity data;
• determine delays for activities in the activity data;
• determine a view value for activities in the activity data based on a corresponding time to data wave migration;
• determine an activity value for activities in the activity data based on a corresponding activity status;
• determine a predicted delay of the cloud migration project based on at least a first iteration and a second iteration of a random forest model on the historical activity data for previously completed cloud migration projects, wherein the first iteration selects an optimum number of data trees in the random forest model, and the second iteration fine tunes model arguments in the random forest model;
• and
• determine a predicted delay score based on a weighted consideration of the predicted delay, the view value, and the activity value.