Proactive Network Management in Cable MSO’s: Data Synthesis and KPI’s – managing a tidal wave of information (Part 4 of 5)
05 April 2016
In our last blog we talked about the importance of having the right data at the right time and how to ensure you are maximizing your efficiency in the collection of this data for operational use. In this post we will focus on how to synthesize the many different data sources into comprehensible results that drive intelligent troubleshooting. Providing simple to use tools to operations and engineering teams improves Mean Time to Repair (MTTR) which directly correlates to an improved customer experience.
Specific measures and Key Performance Indicators (KPI’s) for maximum customer experience
There are a number of different measurements available to determine the health of both the network as a whole as well as the individual CPE devices that are a part of that network. These measurements provide valuable clues in relation to the health and wellbeing of a cable system. In some cases single types of KPI’s or alerts may paint a picture of potential issues and in others it may require a combination of variables and inputs to expose an underlying problem.
Identifying all of the different information that can correlate to a specific type of problem on a physical network is a combination of knowledge of each of these indicators as well as confirmation through uses cases and testing. The addition of more data points and inputs that relate to a particular problem and then presenting it geospatially can provide angles and insights that might not be seen otherwise. We have estimated that providing this data in a single view and giving it a geographic context can reduce issue diagnosis time by 30 – 35%.
In the myWorld application example above, service nodes are color-coded in near real-time based on the aggregate customer account, device status and performance thresholds so it’s instantly obvious where the issues are that need to be looked at immediately. Relevant additional information such as trouble tickets, power supply alerts, network topology, smart routers, WiFi AP’s, and fiber routes are spatially and visually correlated to complete the picture and help with decision support and action plans.
There are a myriad of different ways to identify potential issues out in the field on physical plant and at a boiler plate level simple correlation algorithms are very often utilized. A basic example of this would be outage detection. A long established method in the cable industry for determining line or service node problems has been to set a trigger based on numbers of customers calling in from the same node with similar type issues, typically no signal or severely impacted service.
Correlating these and triggering on a certain number of calls to customer service will invariably bring an issue to a providers attention on a larger scale but lacks in the ability to close the loop on smaller pocket issues and outages where the preset criteria might not be met.
For instance a failed line extender that feeds twelve customers downstream from it before terminating will most likely take some time to generate the three or four necessary customer call-ins to bring attention to the issue. These shortcomings have become more and more prevalent in recent years and as the cable industry has transitioned from an entertainment based television provider to an essential part of people’s lives with telephone service and pervasive high speed data at the forefront. This means differentiated and innovative ways to stay out in front of customer issues have become essential.
Specific KPI’s used to Improve Customer Experience
How is customer experience best improved upon when there are so many products and service riding on the line? The answer lies in finding ways to predict potential customer issues before they arise and KPI’s provide the means in which to do that. In the MSO world KPI’s have long been used for active troubleshooting but now more than ever they are being used to draw attention to areas where there is a degradation track leading to an impending failure which will more than likely result in a poor customer experience. By looking at these KPI’s and combining some of them together an even more granular picture can be drawn of underlying issues that may not be noticed by the customer in its present state but history has shown will eventually be severe degradation or even failure.
Don’t ask us, ask your team
Just ask an Outside Plant or Field Technician if they would benefit from being able to bring together Docsis device performance metrics in concert with leakage/LTE, current trouble tickets, power supply alerts, account status, downstream/upstream data usage, WiFi, fiber, structure, and a myriad of other potential data sets all geographically positioned and visible on top of their network design architecture.
They would be able to demonstrate many ways that these datasets work in concert to converge on root cause issues and paint a more complete picture than if located in different applications. The days of driving out to an area without the benefit of valuable intelligence on where you should be starting your troubleshooting already available and visualized should be numbered if not already gone. And with the additional exposure of true proactive network management data points such as pre-equalization coefficients there is now the added opportunity of being able to visualize underlying issues that eventually lead to node health problems before conventional methods even can show them.