Glossary (Optimization Engine)

Optimization Engine Term	Explanation

Optimization Engine Term	Explanation
AMAS	Adaptive Multi-Agent System is based on the concept of "agents". Its global function is to converge towards a state where user-defined criteria are satisfied. In Pricefx, AMAS is called Multi-Agent Optimization Engine (OE).
Agent	An autonomous program that indefinitely loops its lifecycle of perception-decision-action. “Autonomous” means the agent decides itself what to do, when to do it, and how to do it.
Constraints Parameters	Constraints are threshold and target criteria. Their parameters are namely: Acceptable delta – Acceptable margin of error. Precision – Expected precision of the target. More details are in Criteria Description.
Computation	An agent which takes variables as input to compute a value that it assigns to the output variable. This is one of the concepts of the meta-model.
Coordinates	Associate dimensions (e.g. product, client) to categories (e.g. product X, client Z) to situate variables in the problem's space. This is one of the concepts of the meta-model.
Criterion	It is an agent which expresses a criticality level by reflecting their satisfaction and their importance, e.g. targets, thresholds, or mini/maximization objectives. Criteria are applied to a variable. This term is one of the concepts of the meta-model.
Dimensions	They are what defines levers and criteria. For instance, if your levers are product prices and your criteria are margin targets per product family and customer, then your dimensions are product, product family, and customer.
Dissatisfaction (criterion)	How much a criterion is dissatisfied, high = bad.
Elite	The smallest group of the most influential Value Finders whose cumulated influence is greater than half of the total influence received by the Criterion Agent.
Graph Patterns	Simple computation graph patterns to test the system and identify some Non-Cooperative Situations. They include SISO – Single input single output, SIMO – Single input multiple outputs, MISO – Multiple input single output, MIMO – Multiple input multiple outputs, Asynchronous feedback, Phase shift, Phase interference, Cycle.
Induced Movement (criterion → value finder)	How much the value finder has moved because of the criterion.
Influence (value finder → criterion )	Estimation of how much the criterion’s dissatisfaction has moved because of the value finder’s actions.
Influence proportion (value finder → criterion)	Weight of the value finder’s influence on the criterion, relative to the other influencing value finders.
Impact (criterion → value finder)	It defines how much the criterion pressures the value finder, relative to other neighboring criteria.
MAS	Multi-Agent System which performs multi-dimensional multi-criteria optimization.
Messages	They are the objects used by agents to communicate with each other. Their types are: Notify Value, Notify Criticality, Notify Elite, Subscribe to Value, Set Value, Notify No Longer Critical.
Non-Cooperative Situations	The AMAS approach states that agents must be cooperative: they should seek to help each other, adopt a useful behavior, and communicate useful information in a non-ambiguous form. This is the way to ensure that local behaviors lead to an adequate global function. Some situations may prevent agents to cooperate. They are called Non-Cooperative Situations (NCSs) and are of different types depending on what is going wrong at what part of the lifecycle (e.g. conflict, uselessness, ambiguity, etc.). An agent should be able to either detect and solve NCSs or anticipate and avoid them. This documentation mentions the following types of NCSs: Computation Agent Conflict (Phase Shift), Value Finder Conflict (El Farol Bar Problem), Value Finder Unproductiveness (No Neighbor), Value Finder Incompetence (Asynchronous Feedback), Criterion Agent Conflict (Gradient Loss).
PO.AI	It is the previous name of the Optimization Engine module. PO for Price Optimizer, AI for Artificial Intelligence.
Problem Modeling	Process of understanding the customer’s domain and requirements, and turning them into a problem description file that is fed to the Multi-Agent Optimization Engine.
Problem Description File	A JSON file built from a Groovy map, describing the optimization problem, sent to the Optimization Engine.
Scope (element of problem description file)	Located within a space, it defines the perimeter in which the following variables and criteria exist.
Space (element of problem description file)	Defines the dimensions in which the following variables and criteria exist (i.e. a plate in the graphical norm).
Stamp	Part of a message which helps track the origin of the changes that spread through the computation graph.
Stop Condition	Defines when Multi-Agent Optimization Engine should stop. The types of conditions are the maximum number of steps, the minimum dissatisfaction level, and the minimum entropy level.
Value Finder	It is an agent responsible for finding an adequate value for themselves, i.e. a value that satisfies some user-defined criteria. Typical Value Finders include list prices and discounts.
Value Finder Parameters	Settings for certain types of Value Finders. The settings include initial value, initial/minimal/maximal amplitudes, and increase/decrease coefficients.
Variable	An agent which maintains and exposes a value found in the problem. This is one of the concepts of the meta-model.