Measuring the Mobile Phone Video Service Utility Based on a Service Value Network Model

Similar documents
The Research on Maturity of County Economy Collaborative Development Based on Circular Economy

Doing Business Better. The benefits of becoming a LASA Affiliate

986 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 33, NO. 5, MAY 2015

Optimal Placement of Sectionalizing Switches in Radial Distribution Systems by a Genetic Algorithm

Performance Analysis of an Enhanced DQRUMA/MC-CDMA Protocol with an LPRA Scheme for Voice Traffic

Definition of level and attenuation in telephone networks

Comparative Analysis of Reuse 1 and 3 in Cellular Network Based On SIR Distribution and Rate

Multi-objective Transmission Planning Paper

Systematic Approach for Scheduling of Tasks and Messages under Noise Environment

To: Professor Avitabile Date: February 4, 2003 From: Mechanical Student Subject: Experiment #1 Numerical Methods Using Excel

A Cooperative Spectrum Sensing Scheme Based on Trust and Fuzzy Logic for Cognitive Radio Sensor Networks

A study of turbo codes for multilevel modulations in Gaussian and mobile channels

Decision aid methodologies in transportation

Neuro-Fuzzy Network for Adaptive Channel Equalization

A Non-cooperative Game Theoretic Approach for Multi-cell OFDM Power Allocation Ali Elyasi Gorji 1, Bahman Abolhassani 2 and Kiamars Honardar 3 +

LS-SVM Based WSN Location Algorithm in NLOS Environments

A Data-Driven Robustness Algorithm for the Internet of Things in Smart Cities

Digital Transmission

Formation Control For Wheeled Mobile Robots

LMS Beamforming Using Pre and Post-FFT Processing for OFDM Communication Systems

Efficient Large Integers Arithmetic by Adopting Squaring and Complement Recoding Techniques

PRACTICAL, COMPUTATION EFFICIENT HIGH-ORDER NEURAL NETWORK FOR ROTATION AND SHIFT INVARIANT PATTERN RECOGNITION. Evgeny Artyomov and Orly Yadid-Pecht

The Qualiflex Method For The Insurance Company Selection Problem

Dynamic SON-Enabled Location Management in LTE Networks

Dynamic Optimization. Assignment 1. Sasanka Nagavalli January 29, 2013 Robotics Institute Carnegie Mellon University

Algorithms Airline Scheduling. Airline Scheduling. Design and Analysis of Algorithms Andrei Bulatov

A MODIFIED DIRECTIONAL FREQUENCY REUSE PLAN BASED ON CHANNEL ALTERNATION AND ROTATION

A Preliminary Study on Targets Association Algorithm of Radar and AIS Using BP Neural Network

Cooperative Wireless Multicast: Performance Analysis and Power/Location Optimization

STRATEGIES TO SUPPORT AMBULANCE SCHEDULING WITH EFFICIENT ROUTING SERVICES

UWB & UWB Channels HANI MEHRPOUYAN

CONCERNING THE NO LOAD HIGH VOLTAGE TRANSFORMERS DISCONNECTING

A TWO-PLAYER MODEL FOR THE SIMULTANEOUS LOCATION OF FRANCHISING SERVICES WITH PREFERENTIAL RIGHTS

Uncertainty in measurements of power and energy on power networks

Research of Dispatching Method in Elevator Group Control System Based on Fuzzy Neural Network. Yufeng Dai a, Yun Du b

Analysis of Packet Combining for Single Carrier Multi-Relay Broadband System

Optimal and water-filling Algorithm approach for power Allocation in OFDM Based Cognitive Radio System

Recurrent Neural Network Based Fuzzy Inference System for Identification and Control of Dynamic Plants

A Comparison of Two Equivalent Real Formulations for Complex-Valued Linear Systems Part 2: Results

Journal of Applied Science and Agriculture, 9(4) April 2014, Pages: Journal of Applied Science and Agriculture

Performance Analysis of Multi User MIMO System with Block-Diagonalization Precoding Scheme

A MODIFIED DIFFERENTIAL EVOLUTION ALGORITHM IN SPARSE LINEAR ANTENNA ARRAY SYNTHESIS

aperture David Makovoz, 30/01/2006 Version 1.0 Table of Contents

User Based Resource Scheduling for Heterogeneous Traffic in the Downlink of OFDM Systems

Decision Analysis of Dynamic Spectrum Access Rules

A Tractable and Accurate Cross-Layer Model for Multi-Hop MIMO Networks

29. Network Functions for Circuits Containing Op Amps

The Spectrum Sharing in Cognitive Radio Networks Based on Competitive Price Game

A Multi-standard Efficient Column-layered LDPC Decoder for Software Defined Radio on GPUs

Approximating User Distributions in WCDMA Networks Using 2-D Gaussian

THE third Generation Partnership Project (3GPP) has finalized

Analysis and Enhancement of Bandwidth Request Strategies in IEEE Networks

熊本大学学術リポジトリ. Kumamoto University Repositor

10 9 Channel on time. sec

A Study on Mechanism of the Growth and Evolution of Intellectual Property Value Chain: A Self-Organization Perspective

6928 IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL. 13, NO. 12, DECEMBER 2014

High Speed, Low Power And Area Efficient Carry-Select Adder

A NSGA-II algorithm to solve a bi-objective optimization of the redundancy allocation problem for series-parallel systems

Application of Intelligent Voltage Control System to Korean Power Systems

Optimized Forwarding for Wireless Sensor Networks by Fuzzy Inference System

An Improved Profile-Based Location Caching with Fixed Local Anchor Based on Group Deregistration for Wireless Networks

Proceedings of the 6th WSEAS International Conference on Applications of Electrical Engineering, Istanbul, Turkey, May 27-29,

Impact of Interference Model on Capacity in CDMA Cellular Networks. Robert Akl, D.Sc. Asad Parvez University of North Texas

Distributed Computation in Dynamic Networks

Priority based Dynamic Multiple Robot Path Planning

Equivalent Line Limit Calculation Using Available Thermal Transfer Capability

EE 330 Lecture 22. Small Signal Analysis Small Signal Analysis of BJT Amplifier

UNIT 11 TWO-PERSON ZERO-SUM GAMES WITH SADDLE POINT

Integrity Data Attacks in Power Market Operations

IEE Electronics Letters, vol 34, no 17, August 1998, pp ESTIMATING STARTING POINT OF CONDUCTION OF CMOS GATES

1.0 INTRODUCTION 2.0 CELLULAR POSITIONING WITH DATABASE CORRELATION

The Effect Of Phase-Shifting Transformer On Total Consumers Payments

A New Regressor for Bandwidth Calculation of a Rectangular Microstrip Antenna

Resource Allocation Optimization for Device-to- Device Communication Underlaying Cellular Networks

A Simple Satellite Exclusion Algorithm for Advanced RAIM

Learning Ensembles of Convolutional Neural Networks

Chapter 13. Filters Introduction Ideal Filter

NETWORK 2001 Transportation Planning Under Multiple Objectives

Multi-Source Power System LFC Using the Fractional Order PID Controller Based on SSO Algorithm Including Redox Flow Batteries and SMES

Figure 1. DC-DC Boost Converter

Adaptive Modulation for Multiple Antenna Channels

White Paper. OptiRamp Model-Based Multivariable Predictive Control. Advanced Methodology for Intelligent Control Actions

Topology Control for C-RAN Architecture Based on Complex Network

NATIONAL RADIO ASTRONOMY OBSERVATORY Green Bank, West Virginia SPECTRAL PROCESSOR MEMO NO. 25. MEMORANDUM February 13, 1985

Side-Match Vector Quantizers Using Neural Network Based Variance Predictor for Image Coding

A Tractable and Accurate Cross-Layer Model for Multi-Hop MIMO Ad Hoc Networks

Multiarea Transmission Cost Allocation in Large Power Systems Using the Nodal Pricing Control Approach

ESTIMATION OF DIVERGENCES IN PRECAST CONSTRUCTIONS USING GEODETIC CONTROL NETWORKS

Hyper-cycle Self-organizing Evolution of Industrial Innovation

Coverage Maximization in Mobile Wireless Sensor Networks Utilizing Immune Node Deployment Algorithm

NOVEL ITERATIVE TECHNIQUES FOR RADAR TARGET DISCRIMINATION

MTBF PREDICTION REPORT

Modelling Service Time Distribution in Cellular Networks Using Phase-Type Service Distributions

Development of Analytical Models for Switched Reluctance Machine and their Validation

Spread Spectrum Image Watermarking for Secured Multimedia Data Communication

Walsh Function Based Synthesis Method of PWM Pattern for Full-Bridge Inverter

Channel Alternation and Rotation in Narrow Beam Trisector Cellular Systems

Design and Implementation of a Sort Free K-Best Sphere Decoder

Evaluation of Kolmogorov - Smirnov Test and Energy Detector Techniques for Cooperative Spectrum Sensing in Real Channel Conditions

Chapter 1. On-line Choice of On-line Algorithms. Yossi Azar Andrei Z. Broder Mark S. Manasse

Transcription:

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty Measurng the Mobe Phone Vdeo Servce Utty Based on a Servce Vaue Network Mode Jnuan Ren Communcaton Unversty o Chna renjnun@cuc.edu.cn Bo Communcaton Unversty o Chna bobb@cuc.edu.cn Xaoong u Communcaton Unversty o Chna 3898884@qq.com Cuyng We Communcaton Unversty o Chna wecuyng@63.com hua u Communcaton Unversty o Chna hu@cuc.edu.cn png Zhao Unversty o Manchester png.zhao@manchester.ac.uk Hu u Communcaton Unversty o Chna uhu@cuc.edu.cn Chuanzhen Wang Communcaton Unversty o Chna wangchuanzhen@63.com ABSTRACT Servce vaue networks (SVNs are a new orm o ndustra cooperaton. Wthn SVNs, servce provders engage n oosey couped reatonshps to jonty oer dynamcay congured compex servces to customers. Vaue s co-created through these jont compex servces. By specazaton, servce provders everage the knowedge and capta assets o ther partners, and share the rsk o operatng n a changng and uncertan envronment. However, as a new approach, SVNs ack theoretca oundatons and emprca studes to vadate ts appcabty. Ths paper seeks to address these probems by makng three contrbutons. Frst, based on an exstng theoretca SVN mode, the paper presents a desgn or a mobe phone vdeo servce vaue network (MPVSVN n Chna. Second, we descrbe a systematc approach to cacuate the vaue and utty o servces on the SVN. Thrd, we demonstrate our approach n detaed steps through rea word servce exampes dented rom the MPVSVN. Keywords Servce vaue networks (SVNs, mobe phone vdeo servces (MPVSs, compex servces, servce vaue, servce utty. INTRODUCTION Accordng to Internet Trends pubshed by KPCB (Meeker and Wu,, there were one bon smartphone subscrbers n the word n, representng a % o the entre popuaton o the mobe phone users. In the same year, the smartphone shpment to Chna amounted or 6.5% o the tota goba shpment, makng Chna the bggest smartphone market n the word. It was predcted that by the end o 3, the popuaton o the smartphone users n Chna woud grow beyond 5 mon (amas, Stoega, Drake, Crook, and Homan,. The rapdy growng mobe phone market has opened up many busness opportuntes or servce provders; yet, at the same tme, as more and more peope are usng smartphones, servce provders are aso acng rsng demands and tough compettons or sophstcated, customzed products. In such a dynamc market, tradtona vertca suppy chans can no onger meet the ever-changng demands o the customers and servce provders now seek new orms o cooperaton. One cooperaton orm s caed servce vaue networks (SVNs (Bau, Kramer, Conte, and Dnther, 9. In such networks, servce provders engage n oosey couped reatonshps to oer dynamcay congured jont compex servces to customers. Through specazaton, servce provders everage the knowedge and capta assets o ther partners, and ower the rsk o operatng n a changng and uncertan envronment (Bau, Kramer, Conte, and Dnther, 9. Drven by arge companes such as IBM and SAP (Basoe and Rouse, 8, SVNs are a promsng cooperaton approach and represent the current trend n the servce ndustry. However, as a new approach, SVNs ack theoretca oundatons and emprca studes to vadate ts appcabty. To our knowedge the most compete ormazaton o SVNs to date s gven by Bau, Kramer, Conte, and Dnther (9, and Bau, Wenhardt, Dnther, Conte, and Xu (9. These authors deveoped a orma mode o SVNs and provded a precse denton to characterze what SVNs are and what consttutes compex Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3.

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty servces on SVNs. However, ther work s purey theoretca and has not been vadated by emprca studes. Ths paper seeks to address these probems by makng three contrbutons: Frst, we desgn a SVN or mobe phone vdeo servces (MPVSs n Chna. The purpose o ths desgn s to demonstrate the practca appcabty o Bau et a s SVN mode. Second, we provde a servce utty uncton and based on ths uncton we ormuate a systematc approach to cacuate the vaue and utty o servces on the SVN. Thrd, we demonstrate our approach n deta through rea word servce exampes dented rom the mobe phone vdeo servce vaue network (MPVSVN. Ths paper proceeds as oows. Secton rst denes SVNs and ther reated concepts, and then ntroduces a servce utty uncton. Secton 3 then appes the SVN mode to the desgn o the MPVSVN n Chna. Based on the utty uncton dened n Secton, Secton 4 descrbes and ustrates a systematc method or cacuatng servce utty. Secton 5 dscusses some cosey reated work and nay, Secton 6 concudes the paper. DEFINITION Servce Vaue Networks Vaue and vaue creaton are at the heart o the servce ndustry. In the servce market, vaue s co-created as we as codetermned by servce vaue networks (SVNs. Bau, Kramer, Conte, and Dnther (9 dened a SVN as consstng o a set o servce provders ( s S that suppy a portoo o servce oers ( v V that provde descrbed unctonaty. Each servce provder can own one or more servce oers ndcated by an ownershp reaton. Servces and ther connectons orm a graph-ke structure that s drected and acycc startng rom a source node v and endng at a snk node v. For convenence the terms servce oer and servce can be used nterchangeaby (Bau, Wenhardt, Dnther, Conte, and Xu, 9. s s s s 3 s v v cs c a a a v a vs c 4 v v s v v 3 cs3 c34 a 3 a 3 a 4 v 4 a 4 Y Y Y Fgure. An Abstract Servce Vaue Network Mode (Bau, Wenhardt, Dnther, Conte, and Xu, 9 Fgure shows an abstract mode o SVN wth our servce oers V = { v, v, v3, v4} that are owned by three servce provders S = { s, s, s3}. Substtutabe servces are those provdng roughy smar unctonaty and are custered n canddate poos ( y Y. A canddate poo s a set o potenta servces Y = { v, v3} that can be repaced on-demand. Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3.

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty Interoperabe servces are those havng compatbe nteraces and nput and output capabtes, and such servce oers orm a drected composton reaton. Accordng to Fgure, there are three pars o nteroperabe servce oers: ( v, v, ( v, v 4, and ( v 3, v 4. Servce Vaue, Compex Servces and Servce Conguraton Servce vaue s created through the SVN by perormng a sequence o servces that orm a connected path rom source to snk. Such created vaue s caed a compex servce (Bau, Wenhardt, Dnther, Conte, and Xu, 9. et F denote the set o a easbe paths rom source to snk. Every F represents a possbe nstantaton o the compex servce consstng o unctonaty rom each canddate poo. In Fgure, a compex servce can be nstantated ether by a composton o v and v, or v and v 4, or v 3 and v 4. Servces are characterzed by ther attrbutes (Bau, Wenhardt, Dnther, Conte, and Xu, 9. Servce attrbutes can be ether unctona or non-unctona (e.g. avaabty, prce or securty. Non-unctona attrbutes are aso caed quaty o servce (QoS requrements (Sun, Zhao, oucopouos, and Zhou, 3. Servce provders can congure each servce accordng to ts attrbutes. A servce conguraton V s uy characterzed by a vector o attrbutes (,,..., = where A a a a A o servce a s an attrbute vaue o attrbute o servce s conguraton. The conguraton o a servce represents the quaty eve provded by the servce and derentates ts oerng rom other servces. The conguraton A o the compex servce s the aggregaton o a attrbute vaues o contrbutng servces on the path such that A = ( A, A,..., A wth A = a (Bau, Wenhardt, Dnther, Conte, and Xu, 9. The aggregaton operator o attrbutes vaues depends on ther type (.e. encrypton can be aggregated by an AND operator whereas response tme s aggregated by a sum operator. These concepts w be used n the oowng secton when we ntroduce a utty uncton or cacuatng the servce utty. Servce Utty uncton We dene servce utty U ( A as the overa weghted vaue o a conguraton A o the compex servce represented by path. U ( A s cacuated by a utty uncton as n Formua (. n U ( A = µ U ( ( Where = µ s the weght o compex servce on servce path and n µ =. U ( represents the utty vaue o compex servce on path. Supposng servce has attrbutes, U ( s measured by the weghted average o the vaues o a attrbutes o servce, as shown n Formua (. = ( U ( w a (, = Where a ( represents the th attrbute vaue o each ndvdua servce on path and a (. w= ( w, w,..., w s the weght vector o derent attrbutes n compex servce and = w s the weght or attrbute w =,. In the oowng sectons, we rst produce a desgn or the MPVSVN n Chna and then descrbe a method that uses Formuae ( and ( to cacuate the utty o compex servces or MPVSs n Chna. == Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3. 3

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty DESIGNING A MOBIE PHONE VIDEO SERVICE VAUE NETWORK MODE Based on the SVN mode descrbed n Secton, ths secton produces a desgn o a MPVSVN mode or servce provders n Chna. Accordng to our market research, currenty the MPVS provders n Chna are made o the ve major groupngs: Handset provders that consst o Appe, SAMSUNG, HTC, NOKIA, MIUI, etc. Operatng system (OS patorm provders consstng o Appe, Noka and Mcrosot. Mobe network provders ncudng Chna Mobe, Chna Uncom, and Chna Teecom. Mobe phone vdeo content aggregaton provders ncudng YOUKU &TUDOU, ESHI, QIYI and so on. Mobe phone vdeo content provders ncudng end-users and meda organzatons, such as CCTV and Zhongyng m producer, etc. The MPVSVN n Chna conssts o these servce provders and ther servce oers. Based on the structure and components n the SVN mode (Fgure, we have deveoped a MPVSVN mode n Chna, as Fgure shows. Fgure. The Mobe Phone Vdeo Servce Vaue Network (MPVSVN n Chna Accordng to Fgure, the MPVSVN has ve canddate poos, consstng o Handsets, OS, Networks, Appcatons and Content. Each canddate poo has a set o substtutabe servce oers. For exampe, the Handset poo contans Phone, Samsung, HTC, MIUI, and Noka. Based on the number o canddate poos and the number o servce oers shown n Fgure, the maxmum number o potenta composton paths coud be 5x3x3x5x3 = 676. Exampes o paths are Phone OS Chna Unon etv TV Programs and HTC Androd Chna Mobe PPS User Generated Vdeos. Snce each servce can be congured nto derent conguratons, the number o potenta compex servces can be nnte. The MPVSVN thereore can oer the servce consumers an nnte number o congurabe servces. In the oowng secton, we show how to cacuate the overa utty o compex servces on the MPVSVN. Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3. 4

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty A MUTI-OBJECTIVE OPTIMIZATION APPROACH TO MEASURING SERVICE VAUE AND UTIITY Based on Formuae ( and (, we have ormuated a systematc approach to measurng the servce utty o the MPVSs. Ths approach conssts o our major steps and ther assocated methods, summarzed n Tabe. Step : Determnng the attrbutes and attrbute vaues. Step : Cacuatng the weghts or a the attrbutes. Step.: Constructng the standardzed objectve attrbute matrx. Step.: Constructng the customer preerence matrx. Step.3: Cacuatng the weghts by aggregatng objectve attrbute vaues and costumer preerence vaues. Step.4: Cacuatng the subjectve weghts through expert scorng. Step.5: Aggregatng the objectve and subjectve weghts. Step 3: Cacuatng the servce vaues. Step 4: Cacuatng the overa utty or the compex servce. Tabe. A Systematc Approach to Cacuatng the Overa Utty o A Compex Servce Snce the method presented n Tabe s genera, or brevty, the oowng sectons ony show how to cacuate the vaue or servces n the handset and OS poos. In our cacuaton, we have combned the vaue o servces n the handset and OS poos together because every handset must have an OS as a mnmum compex servce. Step : Determnng the Attrbutes and Attrbute Vaues The purpose o ths step s to denty the attrbutes or a the servce oers n each canddate poo and assgn a numerca vaue to each attrbute. For brevty, here we ony consder attrbutes or the Handset and OS poo n the MPVSVN, whch has ve substtutabe servce oers, whch are Phone 4S, Samsung Gaaxy, HTC VT, MIUI S, and Noka uma. Our market research normed us that the most mportant attrbutes that concern the servce consumers or handset and OS are Prce, Market Share o OS, Screen Sze, Resouton Rato, RAM, ROM, and Battery Capacty. For each o these attrbutes, we cacuated ts average market vaue. Tabe summarzes the set o attrbutes or the handset and OS canddate poo and ther vaues. Attrbute Phone Prce (RMB Market Share o OS (% Screen Sze (nch Resouton (Pxe RAM (MB ROM (GB Battery (mah Phone 4S 4488 3.5 644 5 6 4 Samsung Gaaxy 4699 6 4.8 96 4 4 HTC VT 999 6 4 384 5 4 65 MIUI S 69 6 4 499 4 4 93 Noka uma 499 5 3.7 384 5 6 45 Tabe. The Idented Attrbutes and Attrbute Vaues or the Handset and OS Canddate Poos Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3. 5

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty Step : Cacuatng the Weghts or a the Attrbutes The weght o an attrbute determnes the degree o the mportance o the attrbute. The purpose o ths step s to construct the weght vector w= ( w, w,..., w based on three sets o vaues: ( the vaues rom the standardzed objectve attrbute vaue matrx, ( the vaues rom the consumer preerence matrx and (3 the subjectve vaues rom the expert score vectors. Once the weghtng vector w= ( w, w,..., w s cacuated,the vaue o compex servce x s determned by Formua (3. Z ( w = = w r, =,,, m (3 Where r s the Standardzed attrbute vaue o attrbute o servce s conguraton. (,,..., We appy a mut-attrbute decson method to nd the optma weght vector w= w w w, and cacuate the vaue o each servce. The vaues o a the servces are converted nto a compementary judgment matrx by a near converson uncton, whch s descrbed n Formua (5. The ratona weght vector w= ( w, w,..., w shoud mnmze the devaton between the converted compementary judgment matrx and the consumer preerence compementary judgment matrx. Takng the mnmum devaton as the goa, the optmzaton mode s estabshed n Formua (7. The oowng sub-steps descrbe how to construct ths weght vector. Step.: Constructng the Standardzed Objectve Attrbute Matrx th Assumng there are m substtutabe servces n a canddate poo, denoted by x, =,,..., m, or servce x, ts attrbute vaue s denoted by a, =,,..., m, =,,...,. A the attrbute vaues o these substtutabe servces are stored n a matrx A ( a =. m The vaues or the seven attrbutes n the Handset and OS poos are gven n the oowng matrx. 4488 3.5 644 5 6 4 4699 6 4.8 96 4 4 A= 999 6 4 384 5 4 65 69 6 4 499 4 4 93 499 5 3.7 384 5 6 45 Attrbute vaue matrx A contans rare vaues. To obtan an objectve measure o these vaues, we appy a normazaton uncton as shown n Formua (4 to normaze them. r a, max( a = mn( a, a (4 Where represents the benet attrbutes and represents the cost attrbutes. Formua (3 normazes the attrbute vaues by maxmzng the vaues o the benet attrbutes and mnmzng the vaues o the cost attrbutes. In the Handsets and OS poo, there s ony one cost attrbute, whch s Prce; the rest are the benet attrbutes. Based on Formua (3, we have constructed the oowng normazaton matrx or a the attrbutes n the Handsets and OS poo. Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3. 6

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty R=.367.35.79.6667.5..676.3445.....5..899..8333.467.5.5.7857...8333.4448..5.99.6479.833.778.467.5..695 Usng Formua (3, The vaues o the servces are transormed nto a compementary judgment matrx oowng near converson uncton: bj = ( + z ( w z j ( w = ( + w r w rj = = ( ( = + w r rj b j j (5 = + b =, b =.5, b,, j= {,,..., m}. j B = ( b n n j by the Step.: Constructng The Customer Preerence Matrx Customer preerence s aso an mportant actor that nuences the vaue o servce. To obtan customer preerence data, we conducted an onne survey by askng customers about ther preerence to derent handset and OS. From the responses we seected 38 vad answers. We processed these answers by usng a par-wse comparson on the compementary scae o. to B= bj, where b j represent the reatve mportant.9 and represented the resuts n a compementary judgment matrx ( degree o servce oer x comparng to servce oer x j, where bj + bj =, b =.5, bj,, j=,,..., m. The compementary judgment matrx or the attrbute vaues n the handset and OS poos s gven beow..5.5974.656.659.6795.46.5.5744.5744.677 B=.3744.456.5.5564.5667.34.46.4436.5.5359.35.393.4333.464.5 Step.3: Cacuatng the Weghts by Aggregatng Objectve Attrbute Vaues and Costumer Preerence Vaues The purpose o ths step s to cacuate the vaues o the weght vector w= ( w, w,..., w based on the vaues rom the standardzed objectve attrbute matrx and the customer preerence matrx obtaned rom Steps. and.. We use a near devaton uncton to cacuate the devaton between two compementary judgment matrces B= ( b j and B b ( j m m =, as n Formua (6: b b b ( w ( r r j = j j = j + j = m m m m Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3. 7

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty = (( b w ( r r,, j N j j = (6 In order to get a reasonabe weght vector, the smaer devaton s aways better. In dong so, we estabsh the optmzaton uncton to mnmze the dervaton vaue, as shown n Formua (7: mn G= [( ( ] m m m m j = bj w r rj = j= 4 = j= = s.t. w =, w, =,,..., (7 = Step.4: Cacuatng the Subjectve Weghts through Expert Scorng In ths step, we construct an expert score vectors or the attrbutes. Assumng that there are q experts h= ( h, h,..., h q nvoved n scorng and each expert produces a score vector w, the scores gven by the k th expert to the weght vector are denoted as w = ( w, w,..., w, w =, k =,,..., q. k k k k j= kj To aggregate the scores gven by q experts h= ( h, h,..., h q, we appy the optmzaton mode as represented n Formua (8: q mn G= h ( w w s.t. k= = = k k w =, w, =,,..., (8 Formua (8 has two purposes: Frst t ams at mnmzng the sum o the dstances orm the tota weght vaue w to every expert s weght w = ( w, w,... w, k =,,..., q. Second, t ams at mnmzng the sum o the squares o the tota k k k k devaton between every weght and the weght w ( k=,,..., q; =,,...,. k For our study we have nvted experts to score the attrbutes, but the space mtaton prevents us rom presentng the expert scores here. Step.5: Aggregatng the Objectve and Subjectve Weghts Ths step aggregates both objectve and subjectve weghts to nd the optmzed attrbute weghts or the weght vector w= ( w, w,..., w. The aggregaton s perormed by an optmzaton mode, as shown n Formua (9. mn d + d + d + d s.t. 4 q + + m m j j = j= = k= = [(b w ( r r ] d + d = h ( w w d + d = + k + + d, d (9 Where s the optma uncton vaue o Formua (7, and s the optma uncton vaue o mode (8. Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3. 8

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty As an exampe, to nd the respectve vaues o objectve unctons and, we rst use the standardzed attrbute vaue matrx R and customer preerence matrx B to sove the equaton represented n Formua (7, and then use the expert s weght vectors w to sove the equaton n Formua (8. Fnay we appy and to the optmzaton mode (Formua (9 k to obtan the attrbute weght vector o the servces n the Handset and OS canddate poos: w = (.58,.8,.9,.463,.536,.4,.96. Step 3: Cacuatng the Servce Vaues Based on the weghted attrbute vaues obtaned n Step.4, the vaue o servce The vaues or the ve servces n the Handset and OS canddate poo are stored n vector z : z= ( z, z, z, z, z = (.676,.7366,.5843,.6547,.6 3 4 5 x can be obtaned by Formua (3. Where z, z, z3, z4, z 5 denote the servce vaues or Phone, Samsung, HTC, MIUI, and Noka respectvey. By repeatng Steps -3, we can work out the weghted attrbute vaues or the Network canddate poo. The substtutabe servces n ths poo are Chna Mobe, Chna Unon and Chna Teecom (Fgure. These servces are characterzed by our attrbutes, whch are the Number o 3G Subscrbers, Downnk Speed, Vdeo Costs, and Base Statons. The weghted attrbute vaues or the Network poo: w =(.357,.78,.848,.377. The vaues or the three servces n the Network poo are: z = ( z, z, z = (.874,.955,.673 3 Where z, z, z denote the servce vaues or Chna Mobe, Chna Unon and Chna Teecom respectvey. 3 To cacuate the weghted attrbute vaues or the App and Content canddate poos, we used the Customer Satsacton Index (CSI to ndcate the combned servce vaues or servces n these two poos (u, Ren and u, 3, as gven beow. z = ( z, z, z, z, z = (.54773,.544,.5763,.5338,.593 " " " " " " 3 4 5 z, z, z, z, z represent the servce vaues o Youku, Tudou, PPS, etv, and QIY, together wth ther content oers. " " " " " 3 4 5 Step 4: Cacuatng the Overa Utty or Compex Servces In ths step, we rst use Formua ( to cacuate U (, whch represents the utty o compex servce on path. We then appy Formua ( to cacuate the overa utty U( A (e.g., the weghted servce vaue or any conguraton A o a compex servce on path. To determne the servce weghts n Formua (, we apped the expert scorng method smar to Step.3. As an exampe, the overa utty o a compex servce on path : Samsung Chna Unon QIY can be cacuated by nstantatng Formua (, as oows. 3 = = U ( A µ U ( = µ Z + µ Z + µ Z =.7835 3 5 Where µ = (.39,.4,. and z, z, z represent the servce vaues o Samsung, Chna Unon and QIY respectvey. " 5 Fgure 3 depcts a set o compex servces jonty oered by servce provders on the MPVSVN n Chna and ther servce vaues. Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3. 9

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty Fgure 3. Some Compex Servces oered by the MPVSVN n Chna and ther Servce Vaues. DISCUSSION OF REATED WORK As stated n the Introducton secton, SVNs are a new research area. Ths secton dscusses what has been done n ths area and where exstng work s ackng or can be extended. Some schoars have provded the concept o vaue and the ramework o servce vaue network. Aee ( constructed a vaue network that encompasses not ony the ows o products, servces, and revenue, but aso knowedge and normaton. However the network dd not make a dstncton between goods and servces. Hamton (4 dened servce vaue network as the vaue chan through the commerca coaboratve exbty to provde the servces or products dynamcay, and anayzed the ormaton process o the servce vaue network based on e-commerce. Vargo et a (8 studed vaue and vaue co-creaton rom the servce system and servce ogc perspectve. They descrbed the process o vaue co-creaton through nteracton and ntegraton o resources wthn and among derent servce systems. Bau, Kramer, Conte, and Dnther (9 and Bau, Wenhardt, Dnther, Conte, and Xu (9 ntroduced the ormazaton o servce vaue networks as consstng o a set o servce provders that suppy a portoo o servce oers that provde descrbed unctonaty. They proposed an abstract servce vaue network mode. In ths paper, we have nstantated ther mode to anayze the SVN n Chna s MPVSs and demonstrated the practca appcabty o ther conceptua mode. Some researches ocus on the rues and servce strateges that derent servce provders shoud compy wth. Hozer and Ondrus ( examned derent roes n the mobe appcaton vaue chan and determned how mobe network operators can poston themseves n today s turbuent new market. Jarvenpaaand oebbecke (9 studed servce strateges n mobe teevson or three Itaa companes based on the theory o consumer s experence vaue. They used the perspectve o the benets experenced by customers to provde the nsght nto how supper rms may be abe to ncrease the consumer vaue. However, these proposas have ony ocused on the deveopment o statc modes and provded no emprca research support and agorthms that can be mpemented. Consequenty, the questons as to how to measure the vaue o derent servces and how to create more vaue or customers reman unanswered. In ths paper we have dened a utty uncton or cacuatng the servce utty and demonstrated ths uncton through a rea word case study. Most work on vaue networks has concentrated on the enterprse-eve. For exampe, Basoe and Rouse (8 studed vaue networks or avaton, automobe, reta, heath care and teecommuncaton. They ponted out that actors aectng vaue networks ncude technoogca, soca, and economc pocy, and these actors aect nter- and ntra-organzatona Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3.

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty coaboratons. Ther work has not ocused on the mobe servce ed. Zhang et a (6 and Wu et a (7 studed the vaue chan and the vaue network o the mobe busness. Karhu (7 consttuted a vaue network or the mobe teevson servce. (9 proposed that customer vaue promotes ndustra vaue o mobe busness and anayzed the vaue ormaton mechansm n Chnese mobe commerce ndustry. A-Debe and Avson ( dented severa man mobe busness mode dmensons aong wth ther nterdependences and showed that these dmensons contrbuted to a new boundary-ess gobazaton andscape. Very tte eort has been made on vaue cacuaton and anayss. Chen et a (7 proposed a 3V3D nnovaton mode, and showed how to create, mprove and obtan the vaue through vaue proposton, vaue depoyment, and vaue appropraton. They studed the processes o desgn, deveopment and devery o new servces at the enterprse eve, and estabshed the reatonshp between the servces through a busness mode. However ths work was carred out n the context o enterprse vaue networks, rather than the vaue networks that concern both servce provders and consumers. Our method nvoves sovng a mut-attrbute decson probem (MADP. Ths probem has severa soutons. For exampe, Xu (3, 4 presented an optmzaton approach based on a compementary judgment matrx obtaned rom customer preerences. Zhang, Qu and Xn (8 deveoped an optma decson mode based on the tradtona dea pont. So ar, however, tte work has been done to synthesze the objectve actors and subjectve actors such as consumer preerences and expert judgment. In ths paper, we have ntroduced a mut-objectve optmzaton method, whch ntegrates the objectve actors and subjectve actors eectvey by combnng the preerences o consumers and the natura attrbutes o servces. In summary, athough work on mobe servces based on vaue networks has aready begun, exstng research has been many conned wthn enterprses, wth tte attenton to the reatonshp between the mobe servce provders. Our vew s that the vaue network mode shoud be constructed based on the mobe servce provders. Ths paper has presented a pot study o the SVN rom the servce provders perspectve and deveoped an approach to cacuatng the servce utty based on a specc SVN the MPVSVN n Chna. CONCUSION Ths paper seeks to address some practca probems n current SVNs research. The contrbuton o the paper s three-od: Frst, based on an exstng theoretca SVN mode, we have produced a desgn or the MPVSVN n Chna. The desgn has demonstrated the practca appcabty o the SVN mode. Second, we have ntroduced a servce utty uncton and based on ths uncton we have ormuated a systematc approach to cacuate the vaue and utty o servces on the SVN. Thrd, the paper has demonstrated our approach n detaed steps through rea word servce exampes dented rom the MPVSVN. Overa, the paper has made an mportant contrbuton to the work n SVNs. Our approach, whe genera, has suered rom two major mtatons. Frst, ts cacuaton methods are too compex and tme consumng. In the paper, we have ony used these methods to cacuate one compex servce. An automated sotware too s needed to cacuate the utty or a potenta compex servces. Second, athough our approach has empoyed a mutobjectve decson support method or optmzng the attrbute vaues, ths method has not been tested n the rea ndustry settng. To address these mtatons, our uture research w seek to deveop computer-asssted toos to support our approach. In addton, we ntend to study and denty recurrng compex servce patterns and ntegrate such patterns n the SVNs to maxmze the overa vaue. Fnay, we w mprove our approach through more rea word ndustry case studes. ACKNOWEDGMENTS We wsh to thank two anonymous revewers and the mn-track char or ther hepu comments. Our grattude aso goes to Proessor Hao Shen or hs knd hep wth cacuatng CSI o vdeo cents. The work reported n ths paper s supported by the Natona Natura Scence Foundaton o Chna, Grant No. 774. REFERENCES. A-Debe, M. M. and Avson, D. ( Busness mode requrements and chaenges n the mobe teecommuncaton sector, Journa o Organzatona Transormaton and Soca Change, 8,, 5-35.. Aee, V. ( Recongurng the vaue network, Journa o Busness Strategy,, 4, 36-39. 3. Basoe, R. C. and Rouse, W. B. (8 Compexty o servce vaue networks: Conceptuazaton and emprca nvestgaton, IBM Systems Journa, 47,, 53-7. Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3.

Ren et a. Measurng the Mobe Phone Vdeo Servce Utty 4. Bau, B., Kramer, J., Conte, T., and Dnther, C. (9 Servce vaue networks, Proceedngs o Commerce and Enterprse Computng (CEC 9, Juy -3, Venna, Austra, IEEE Press, 94-. 5. Bau, B., Wenhardt, C., Dnther, C., Conte, T. and Xu. Y. (9 How to coordnate vaue generaton n servce networks, Busness & Inormaton Systems Engneerng, 5, 343-356. 6. Chen, Y. G., Hseh, P. and ee, C. (7 An ntegrated ramework or managng servce nnovaton, Proceedngs o the PICMET, August 5-9, Portand, Oregon, USA, IEEE Conerence Pubcatons, 696-7 7. Hamton, J. (4 Servce vaue networks: vaue, perormance and strategy or the servce ndustry, Journa o Systems Scence and Systems Engneerng. 3, 4, 469-489. 8. Hozer, A. and Ondrus, J. ( Mobe appcaton market: a mobe network operators perspectve, ecture Notes n Busness Inormaton Processng, 5, 86-9. 9. Jarvenpaa, S. and oebbecke, C. (9 Strategc management mpcatons o a consumer vaue perspectve on Mobe TV, Journa o Inormaton Technoogy, 4,.. Karhu, P. (7 Emergng mobe servce nnovaton markets: the case o the Fnnsh mobe TV servce market, Dssertaton o the Unversty o St. Gaen.., H. (9 M-commerce vaue reazaton mechansm research based on customer vaue, Doctora dssertaton o Jn unversty.. u,., Ren J. and u X. (3 Constructng and anayzng customer satsacton ndex o mobe phone vdeo Servce, Proceedngs o 3 Internatona Conerence on Management Scence & Engneerng, Juy 7-9, Harbn, Chna. 3. amas, R. T., Stoega, W., Drake, S. D., Crook, S. K. and Homan, B. ( Wordwde smartphone mobe OS 6 orecast and anayss, http://www.dc.com/getdoc.jsp?contanerid=38366. 4. Meeker, M. and Wu,. ( Internet Trends. http://www.kpcb.com/nsghts/-nternet-trends-update. 5. Sun, J., Zhao,., oucopouos, P., and Zhou, B. (3 Qra: A quaty requrements anayss approach or servce systems. Proceedngs o Internatona Conerence on Servces Computng, Accepted, 6. Vargo, S.., Mago, P. P. and Akaka. M. A. (8 On vaue and vaue co-creaton: A servce systems and servce ogc perspectve, European Management Journa, 6, 45 5. 7. Wu, X., Chen, Q. and Sheng,. (7 Industry vaue chan research o the mobe commerce, Journa o Chongqng unversty (soca scence edton, 3, 6, -6. 8. Xu, Z.S. (3 A method or mutpe attrbute decson makng wthout weght normaton but wth preerence normaton on aternatves, Systems Engneerng Theory and Practce,, -3. 9. Xu, Z.S (4. Method or mut-attrbute decson makng wth preerence normaton on aternatves under parta weght normaton, Contro and Decson,, 9, 85-88.. Zhang, F.W., Qu, S.Y. and Xn, C.X. (8 A knd o mut-attrbute decson makng mode wth preerence normaton and ts appcaton, Statstcs and Decson, 9, 57-58.. Zhang, X. and Wu,Y. (6 Teecom ndustry: rom the decomposton o the vaue chan to the constructon o vaue network, Scence and Technoogy Management Research. 4, 7-74. Proceedngs o the Nneteenth Amercas Conerence on Inormaton Systems, Chcago, Inos, August 5-7, 3.