User:StrategyGameChanger/sandbox

Preemptive
From Wikipedia, the free encyclopedia.

''This article is about a term used in support and operations of telecom networks. For other uses of the word, see Preemption (disambiguation).''

Preemptive (services)
Preemptive (also spelled " Pre-emptive ") is an action to preempt will preclude an event from occurring. To act preemptively, is to foresee a potential threat and act to forestall or prevent its occurrence. Serving or intend to preempt or forestall something, especially to prevent attack by disabling the enemy, “a preemptive strike”.

In telecommunications, Preemptive is an action whose primary objective is (1) with firm determination take an action against an imminent or anticipated network disturbance, (2) a preemptive action must come as a reaction to an imminent condition that is both credible and immediate, and (3) a change implemented to address a weakness that is not yet causing a non-conforming product, solution, or service condition, and corrective actions not yet visible through active alarms. Preemptive actions are carried out with deep product-near insights and expertise to ensure appropriate expedient actions while minimizing harm.

Preemptive examples in telecommunication
Preemptive activities have helped best-in-class organizations to take cost control and customer experience to new levels, building trust and customer loyalty.

In telecommunications, Preemptive actions leverage product near data collection, and optimized machine learning, combined with deterministic analytics decision making in near real-time to ensure appropriate quick actions while minimizing false positive implications. A preemptive action effectively prevents an underlying cause of an imminent network anomaly, going beyond only addressing its symptom(s). When taking preemptive action, speed and precision is of essence, to address and analyze the appropriate data insights.

A preemptive action is carried out with deep domain insights and domain expertise to make the appropriate preemptive action to resolve the cause of an imminent condition without causing further harm, increased operational risk, or acting on false positive indicators. Hence, preemptive actions must be prescriptive, and deterministic, and speed is of essence. The preemptive outcome effect is a sustained positive service continuity and performance effect on the targeted product(s) behavior. Due to the product-near nature of preemptive actions are typically not multi-vendor focused where greater weight is put on symptomatic relief based on predictive analytical factors. Prescriptive analytics solutions use optimization technology to solve complex decisions with millions of decision variables, constraints and trade-offs.

A preemptive approach helps service providers to adopt data driven artificial intelligence powered network operations to ensure service continuity, by identifying and addressing anomalies before they impact network performance or customer experience. Anticipating the actions needed and actively data mining trends, rather than just reacting, require adoption of advanced analytics techniques and demonstrated deeper understanding of the deployed products and solutions, and its product near data, and effects of any prevailing network conditions. Anomalies may have distributed causes (e.g. a set of causes that contribute to the symptoms) in which the identification of underlying causes is probabilistic in nature when running diagnostic analytics, in which case one need to address some variables in relation to each other to address the symptoms. Depending on the operational risks, final determination of appropriate actions is often based on firm deterministic decisions (e.g. identifying a single root cause to take a “deterministic” action).

Unlocking useful data points where preemptive actions can impact customer outcomes require close collaboration between customer operations and domain expertise to be successful. The same approach can help preempt unwarranted product returns or exchanges through problem resolution of causes.

The emergence of Industry 4.0 will further require adoption of data-driven preemptive analytics for maintenance and repairs ahead of disruptions in the surrounding ecosystem for production, to maximize up-time, and improve production throughput.

Preemptive techniques can be a bridge toward product self-healing. It is generally expected that captured AI/ML based learnings product-near analytics from preemptive techniques will be embedded into the evolution of preemption and zero touch capabilities within vendor products. Increased up-take of preemptive sets an accelerated path and link toward self-healing.

Predictive trend indicators may also selectively be used to enrich the insights for preemptive actions where conditions of false positives are clearly understood and managed with extensive AI learning based on past events. Predictive techniques are suitable complements for network level conditions were symptoms help enrich the data points.

Preemptive vs predictive
Preemptive capabilities unlock enabling actions to resolve problems with speed and precision to address the cause of an imminent condition. The term “preemptive” is sometimes confused with the term “preventive”, which is an action to address a potential issue of a foreseeable condition in the future while not currently seen as an imminent condition. Preventive actions are often carried out through scheduled maintenance and regularly repaired machine parts to prevent future downtime.

Predictions may enrich the insights to indicate in advance based on an observation, experience, or scientific reason, with some probability, that unconditionally something will happen. Predictive analytics help predict conditions of upcoming failure by synthesizing operational, product and environmental data, enabling actionable insights for improved corrective actions. However, reliable predictive analytics relies on volumes of clean data for training or otherwise risk acting on false positive outcomes. Predictions can be multi-vendor focused where greater weight is put on symptomatic relief based on predictive analytical factors.

Depending on the operational risks, final determination of appropriate actions is often based on firm deterministic decisions. Similarly, proactive actions are intended initiate change rather than reacting to events, to cause changes of preventive effect rather than just reacting to change. To be proactive, you perceive the facts clearly and take constructive action.