Home / Technology / AI & IT / Military decision support require efficient optimization algorithms

Military decision support require efficient optimization algorithms

Research in military mission planning is  an important field. Today military operations rely on increasingly complex joint and multinational environments. This calls for innovative concepts, doctrine, and technologies to support the emergence of new planning and execution systems that are more flexible, adaptive, inter-operable, and responsive to a time-varying and uncertain environment.


The ability to conduct joint operations imposes shared information and interoperability requirements to operate among coalition members. The growing complexity and required transparency impose transit from a rigid vertical organizational structure to a more integrated, modular and tailored one.

In that regard, Network Centric Operations (NCO) offers a unique setting to take on emerging challenges. NCO reflects the true technical transition towards networked action in a machine to machine fashion, with possibilities to host enough computer power to solve hard problems. Almost all types of military equipment have the capability { from mobile phones to ground control stations. Computer power can be distributed as well as decisions.


Another concept that recently has obtained great acknowledgement is Effect Based Operations (EBO) which combine military and non-military methods to achieve a certain effect.  Researchers have also defined EBO in a more narrow context and establish the de nition Effect Oriented Planning (EOP) which basically means that a planning process is primarily focused on the desired effect and evolving backwards from there.


Military requires operational decision support for developing mathematical models of real decision problems, using cutting-edge optimization algorithms to solve these problems, and designing and delivering prototypical decision support tools to the appropriate commands that are tasked with solving these problems.


Some of these problems are Navy multi-ship multi-mission planning for large combat operations, unmanned aerial system planning and assignment, logistics, weapon-target paring, missile-target assignment, theater ballistic missile defense,

Mission planning

The mission planning and operations process is of course more than just planning. In a mission cycle, intelligence must be included in order to always have a clear and updated situation picture. In the execution phase coordination and re-planning in real time is crucial in order to achieve mission goals, however this is complicated and challenging in terms of fi nding the supporting methods and models. Clearly to carry through all steps in mission planning and execution, a loop can be stated that can be repeated over and over again. This operational loop, the OODA loop, was invented by Colonel John Boyd, a United States Air Force and a Pentagon consultant.


The OODA loop gets its name from its four primary steps:

Observe A continuous scanning and collection of data in order to assess a state of threat or anomaly.

Orient If an antagonistic intent is detected an analysis and synthesis of data has to be performed in order to clarify intent and ones current perspective.

Decide The determination of a course of action based on ones current perspective

Act The physical action of those decisions.


Boyd emphasized that this is not a closed loop but rather a series of loops, never-ceasing. He theorized that successful large entities had a series of OODA loops at tactical, operational art, and strategic levels and that the most successful organizations had a highly decentralized chain of command that utilized directive control in order to better utilize the creative and intellectual abilities of individuals at all levels. Whether in aerial combat or in business, the same theory and logic applies: Observe, Orient, Decide and Act.


Planning Hierarchy

Military forces carrying out previously described EBO and necessary planning and execution within the OODA loop, can be divided into three hierarchical levels, namely, strategic, operational, and tactical. Each level of planning corresponds
to a level of conflict.

1. The strategic level of a conflict is that level at which a nation or group of nations determines national or alliance security objectives and develops and uses national resources to accomplish those objectives. Activities at this level establish strategic military objectives, defi ne a desired end state, sequence the objectives, de fine limits and assess risks, and other capabilities in accordance with strategic plans. At the strategic level, different tools and OR techniques can be used. However, since this top level holds a lot of soft values hardly measurable, mostly manual and interactive type of \war gaming” processes dominates. Risk assessment and end state achievements can be approached by Bayesian techniques in combination with manual processes.


2.  The operational level of a conflict is the level at which campaigns and major operations are planned, conducted and sustained to accomplish the strategic objectives. Activities at this level link tactics and strategy by establishing the necessary operational objectives. In order to nd out the proper course of action (COA), planning of operations is crucial but an
extremely complicated task. From an OR perspective, planning tools at an operational level have a higher grade of model based content. Simulation based assessments for operational outcomes can be used together with Artifi cial Intelligence (AI) methods, also in inuence diagram and Petri-Nets are suitable for sequencing issues. Mathematical programming is a well
explored tool in applications such as target clustering, partitioning and force deployment.


3. The tactical level of a conflict is the level at which battles and engagements are planned and executed to accomplish military objectives assigned to tactical units. Activities at this level focus on the ordered arrangement and maneuver of combat elements in relation to each other and to the enemy to achieve combat objectives established by the operational level
commander. At the strategic and operational levels, planning is more of a formal process. At the tactical level, time is often crucial, so fast response is highly regarded. One important objective is to initiate plans and actions within the time-frame of an enemy’s decision cycle. By doing so, the decision maker forces the enemy to become reactive rather than proactive,
which is a huge achievement. OR techniques in tactical planning stems from AI methods such as Genetic Algorithms, mathematical programming techniques and heuristic approaches. Since the task is to develop a COA of a \good enough” quality, one has to choose technique based on the proper trade-o between complexity, accuracy and execution time.



Toshiba’s breakthrough algorithm realizes world’s fastest, largest-scale combinatorial optimization

Toshiba Corporation has realized a major breakthrough in combinatorial optimization—the selection of the best solutions from among an enormous number of combinatorial patterns—with the development of an algorithm that delivers the world’s fastest and largest-scale performance, and an approximately 10-fold improvement over current methods. Toshiba’s new method can be applied to such daunting but essential tasks as identifying efficient delivery routes, determining the most effective molecular structures to investigate in new drug development, and building portfolios of profitable financial products.


The newly developed technique, the Simulated Bifurcation Algorithm, quickly obtains highly accurate approximate solutions (good solutions) for complex large-scale combinatorial optimization problems—problems that have resisted solution for a long time, and that are very difficult to solve using conventional techniques. Potentially even more important, the algorithm also realizes excellent scalability at a low cost using current computers, which could revolutionize current optimization processes.


Toshiba will use the Simulated Bifurcation Algorithm to build a service platform able to quickly solve diverse social and business problems, aiming for commercialization in 2019. Details of the new technology are published in the online academic journal Science Advances.


Many problems can only be solved by sifting through a vast number of options to find the best combinations. These include realizing efficient logistics (the traveling salesman problem in math), directing traffic to ease congestion, applying molecular design to drug development, and optimizing financial portfolios. Today, realizing such combinatorial optimization requires an enormous amount of computation, and using current computers to find solutions remains difficult.


There are growing expectations that next-generation computing devices, such as quantum computers, will lead the way to better solutions, and current research aims to develop computers specially designed for combinatorial optimization through the use of superconducting circuits, lasers, and semiconductor-based digital computers. Despite these efforts, it remains a challenge to increase the solvable-problem size and to reduce the computation time.


For example, it is still difficult for quantum computers with superconducting circuits to solve complex large-scale problems. And while today’s semiconductor-based digital computers have made it easier to increase the solvable-problem size, current algorithms for combinatorial optimization are difficult to parallelize, making it hard to use parallel computing to speed up problem solving.


Toshiba has solved these issues by developing a novel combinatorial optimization algorithm, the Simulated Bifurcation Algorithm. It is highly parallelizable, and can therefore easily speed up problem solving on standard digital computer through parallel computation. As current large-scale computational systems can be used as is, there is no need to install new equipment, making it easy to scale up at a low cost.


For example, by using field-programmable gate arrays (FPGAs), a good solution to an optimization problem with 2,000 fully connected variables (approximately 2 million connections) can be obtained in just 0.5 milliseconds. This is approximately 10 times faster than the laser-based quantum computer recognized as the world’s fastest can solve the same problem. In addition, using a cluster of eight GPUs, Toshiba obtained a good solution for a large-scale problem involving 100,000 fully connected variables (about 5 billion connections) in only a few seconds. These results open up new ways of solving large-scale combinatorial optimization problems in many different areas of application.


The Simulated Bifurcation Algorithm harnesses bifurcation phenomena, adiabatic processes, and ergodic processes in classical mechanics to rapidly find highly accurate solutions. Toshiba derived the principle from a theory of a quantum computer proposed by the company itself. This discovery in classical mechanics inspired by quantum mechanics is an academically interesting, highly novel result that suggests the existence of unknown mathematical theorems.





About Rajesh Uppal

Check Also

Artificial Intelligence (AI) to Enhance Military Intelligence: A New Frontier in Defense Technology

In the rapidly evolving landscape of defense technology, artificial intelligence (AI) is emerging as a …

error: Content is protected !!