MCL is designed for modeling multiple overhead/cathead cranes. It helps to:

• Model interaction of multiple cranes that share common rails
• Model crane maintenance, breakdowns
• Choose the best set of crane operation priorities for your production system
• Increase model development speed

To implement MCL in your project you need to:

1. Create rails, determine the zone, where the cranes can move (x-axis), number of cranes
2. Set sizes of cranes, reach tolerance and «home locations» for cranes
3. Set operation durations for cranes (fix, lift, lower)
4. Determine a set of operations (name, pointA, pointB, priority (double))
5. Set the properties for breakdowns (kinds of breakdowns, rates, durations)
6. Set the maintenance schedule (final users can change it in Excel)

1. Purchase license for the Library (current release). The license is perpetual, you can use it in multiple projects
2. Purchase the technical support (gives you updates and technical support)

Note: Tailored version of the library can be developed according to T&M basis.

The library is based on the optimized loops principle.

At the stage of route planning, checks are being made to determine the presence of the A -> B path.

If such a path exists, the algorithm selects the shortest path, according to minimum possible number of maneuvers and distance traveled.

When the route preparation is done, the train starts moving in compliance with pre-prepared parameters (default speed, indentation, etc.).

If the route is not possible (Example : the length of the train does not allow maneuvering on the only existing shortest path in order to get to the final destination) - the "TrainMove" block will return an error saying moving is impossible and highlighting "problematic" paths.

The library provides connection of various AI algorithms to the model just in one simple action. Depending on the requirements, it is possible to integrate into the model algorithms for searching patterns, machine learning and neural networks.

The use of artificial intelligence algorithms in practice is an advantage that allows achieving previously unattainable speed and accuracy of data processing.

MLC is a bridge between the most popular AI algorithms and the AnyLogic software product.

Classifiers
Classifier - based on the training sample, it compiles an algorithm that allows you to get an arbitrary object from the original set.

Neural networks
Neural networks - a system of connected and interacting simple conditional operators (neurons), allows you to track the complexity and importance of the incoming information.

Linear regression
Linear regression - a method of restoring dependencies between variables.

Decision trees
Decision trees - a way of representing rules in a hierarchical, consistent structure, where each object corresponds to a single node that provides a solution.

Python scripts
If the above mentioned functionality is not enough it is possible to use TensorFlow, Keras, etc. using Python.

MLC also includes a block that allows you to fully execute Python scripts.

Random scenarios where the data flow can vary over time :

Cell towers capacity distribution
Optimal price recommendations while launching a new product based on competitor's parsed data
Energy management solutions based on intelligent equipment operation mode
Market ready product quality forecast
And many more

DMB allows you to write data from the model to the database at any stage of the simulation, one-time or regularly, with a given frequency.
DMB was developed in accordance with requirements for interacting with databases from the Anylogic development environment and has an adapted visual interface that allows you to use databases even with poor knowledge about SQL queries. A key feature of
DMB is transferring the results on the go while the simulation process is running. This provides real time analysis of the output and thus solves the problem of slow and deadlocked connections and allows you to use AnyLogic in conjunction with BI systems.