Impact of Poor Quality in Cellular Networks: a case study of cellular operators in Pakistan

Abid Riaz abidr08@yahoo.co.ukMuhammad Waqar waqar.telenor@yahoo.comMuhammad Adil Shah adilshah57@yahoo.comIshtiaq Malik ishtiaq.fast@gmail.com

Prof. Ali Sajid, Ph.D, (TI) ali@case.edu.pkDirector Strategic Affairs

CASE (Center for Advanced Studies in Engineering), 19 Attaturk Avenue Sir Syed Memorial Building, G 5/1, Islamabad, Pakistan.

Background

Telecom SPs (Service Provider) witnessed a tremendous growth in business over the last decade in Pakistan. This exponential growth was associated to the customer’s attraction to this new technology as the limited mobility concept was about to be history.

But now we saw a drift in customer’s attitude, they are more focused on quality of service. Customers started switching different operators to get better QoS. As a result SPs are loosing the market share and observing the loss in revenues.

The cellular operators analyzed the situation and came to a conclusion- Quality. It was time to slow down the pace of expanding the network and concentrate on improving the quality concerns in each and every aspect of the organization.

Customer retention can be affected by network downtime. If a company is unable to support its customers the loss in confidence may cause them to leave. If this happens too frequently the loss of business will affect revenue and ultimately the stability of the company.

QoS - Cellular Network

The QoS (quality of service) perceived by the customer depends on the performance archived by each component installed in the network and the service resources (end- end).

The service quality can be measured by defining Key Performance Indicators (KPIs) & Key Quality Indicators (KQIs) most relevant to cellular network & services. SPs can never achieve customer satisfaction without properly aligning KPIs & KQIs in a way which truly determines customer requirements.

KQIs are typically combination of several KPIs; the prudent SP will choose to combine KPIs into KQIs to ensure customer satisfaction.

KPI KPI

KQI

Costumer Satisfaction

KQI

KPI KPI

Network Availability

TCH, DCR, CSSR…

Network Downtime

Downtime or outage duration refers to a period of time that a system fails to provide or perform its primary function. This is usually a result of the system failing to function because of an unplanned event, or because of routine maintenance.The company lose revenues due to network element failures. This forces savvy companies to proactively anticipate and prepare for problems before they occur. We propose a model in which the cost of network element failures can be estimated to assess their impacts to companies in terms of failed services and revenue leakage.

Unplanned downtime occurs because of system failure (for example, a hardware failure or a system failure). In contrast, planned downtime is scheduled (hardware or software upgrade); Network administrators can plan for it to occur at a time that least affects customer. In this paper impact of downtime in BSS sector would be tried to quantify to nearest approximation.

Here we can define Network availability as % Availability = (Up Time / (Up Time + Down Time)) ×100

Causes of Downtime (Outages)

1. Unplanned Downtime

Environmental issues Environmental issues include power outages, Air Con failures, power supply/genset failures, rectifier fault, fires, and other disasters.

Hardware issues Hardware issues include memory failures, network card failures.

Media issues   Media issues include link failure, cable damages or loose patching. Software issues   Software grades or failures can reduce network availability. Service issues   Service issues include slow response time & incapable data

center to resolve the outage.

Staffing issues    unqualified or insufficient staff can cause unnecessary delay in outages and lengthen the duration to fix the issue & restore network.

2. Planned DowntimeUpgrades for hardware components & software (usually 10-20% of network downtime).

Measuring the Cost of Network Downtime

1. Assessing the Financial Impact of Downtime

The analysis of outage report of SPs in the 1st quarter of year 2009 revealed the average of different type of outages as under; 83% of environmental issues are due to power supply problems.

Unplanned Outage Distribution

17%

68%

2%10% 3%

Media Issues

Environmental

Hardw are

Access/Weather

Autorestore

The number of affected sites & the average downtime of outages in 1st quarter of year 09 (week 1 – 14) are

Affected Sites

0

100

200

300

400

500

600

w k-01

w k-02

w k-03

w k-04

w k-05

w k-06

w k-07

w k-08

w k-09

w k-10

w k-11

w k-12

w k-13

w k-14

Avg. Dow ntime (Hr)

0:00:00

0:28:48

0:57:36

1:26:24

1:55:12

2:24:00

2:52:48

3:21:36

3:50:24

4:19:12

4:48:00

Wk-01

Wk-02

Wk-03

Wk-04

Wk-05

Wk-06

Wk-07

Wk-08

Wk-09

Wk-10

Wk-11

Wk-12

Wk-13

Wk-14

The trend of average traffic observed during this quarter of the year 2009 is

Avg. Netw ork Erlangs per Week

21.5

22

22.5

23

23.5

24

w k-1

w k-2

w k-3

w k-4

w k-5

w k-6

w k-7

w k-8

w k-9

w k-10

w k-11

w k-12

w k-13

w k-14

x 1

,000,0

00

The simplest way to project the potential annual revenue loss from downtime is with the equationLost Profits = Affected Sites x Duration x Erlangs loss (k)

Let N = number of Affected sites per weekD = Avg. Duration of downtime per week (Downtime per week/No of outages)Er = Erlangs loss per site per hour(Avg. Erlangs per week/No of sites in any network = Avg. Erlangs per week per site)Where coefficient k converts erlangs into revenue (k = 72)

Lost Profit = ∑ i → j (Ni x Di x Eri x k), Where i = week (wk 1, wk 2 ... wk 14, j= 14)

2. Rework/RestorationNormally the major part of environmental issues and to some extent the media issues appear due to non conformance of standards in design, fault rectification & lack of serious efforts in preventive maintenance.

3. Customer ChurnChurn refers to the cost of lost business due to the customer’s dissatisfaction over QoS; it describes the loss of customers.

4. Damaged ReputationDowntime/Service Degradation can also cause a loss in company reputation and customer loyalty.

5. Losing MarginSPs lose significant revenue & margin due to inferior quality; the attitude of CI can help SPs capture quality-sensitive customers.A survey conducted clearly concluded that network reliability and performance are more important than price. The survey of business customers found that 71% of respondents rated availability & performance as important or very important for selecting SP

Improving Network Availability

Operation & Maintenance (O&M) team faces many challenges to increasing network availability. As discussed the downtime can come from planned maintenance activities, unplanned hardware or software failures. However while classifying the downtime SPs not present in reports the factor of human error involved. Usually, a significant portion of outages or delays in fault handlings are associated to human errors or incompetencies.

O&M approach towards outage/fault handling needs classification & evaluation, to develop a strategy to improve network availability in a cost effective way.

The reactive approach promises fast response times to critical problems. All necessary resources will be committed to resolving the situation. Software patches will be implemented to resolve problems if necessary.

The proactive approach focuses on anticipating and addressing potential problems early. A proactive notification of diagnostic indicators can reduce the length of outages, or even eliminate some network outages. Early warning scripts can catch minor issues before they become major ones, giving network operators a wider range of choices and more time to respond, effectively reducing the frequency and duration of network outages.

By inspecting the health check of network nodes on periodic basis the proactive approach defines new dimension and can be quite successful to improve the network availability. Without the attitude of CI (continuous improvement) the proactive approach can not be seen in its real essence.

Conclusions

For the last few years, SPs have focused their energies on heavy roll out, with less emphasis on quality and reliability. The next phase promises SPs survival only if they meet new level network availability & performance. CI in this direction ensures SPs financial benefits too great to ignore & perhaps the only way to survive in this fast growing market having many potential competitors.

“Ounce of prevention is worth a pound of cure"

References

[1] The Impact of Network Downtime on Business Today – Networks first[2] Understanding Downtime - Business Continuity Solution Series™[3] THE BUSINESS CASE FOR DISASTER RECOVERY PLANNING:

CALCULATING THE COST OF DOWNTIME – Iron Mountain[4] Reducing Network Cost - Infonetics Research[5] Measurement and Analysis of End-to-end Service Quality of 3G Networks and

Services – AlbatrOSS[6] What’s Behind Network Downtime? – Juniper

[7] Assessing the Financial Impact of Downtime – Vision Solutions[8] Performance Evaluation of GSM Mobile System in Nigeria.[9] Faults and Service Modeling for Cellular Networks[10] Model Based Diagnosis for Network Communication Faults[11] Prediction of Faults in Cellular Networks Using Bayesian Network Model[12] User Service Performance – Ericsson (White Paper)[13] Rapidly deploy new services with confidence in service quality. - Tivoli[14] Major cellular operators in Pakistan

The paper was presented in the 12th International QMOD and Toulon-Verona Conference on Quality and Service Sciences (ICQSS), 2009.