Transcript Services selection

The impact of service pricing models on service selection
Kristjan Košič
Institute of Informatics
University of Maribor





Service costumers mostly buy results
Clients should be able to control total costs of
services thru scheduled payment schemes
(service to service provisioning)
Service oriented computing and SLA are
drivers of dynamic pricing mechanisms
Adaptive business models
Service offerings and service selection
methods

Impact of service pricing models on service
selection (current main goal)

Dynamic pricing models for e-business

Dynamic: dynamic service behavior, which is
driven by consumers decisions and selections in
the business process

Customers decisions can be guided with the
right information and offer

Overview of existing dynamic business models in
e-business domain

To build taxonomies of business models and
map them to taxonomies of value-creation

Overview and analyze methods for dynamic
price estimation

To build a simulation engine for dynamic service
pricing framework to study and simulate impact
of service pricing models

Learning curve – a simulation based approach to dynamic pricing
[Di Micco / 2003] – product based

Categorization of service e-business models [Narahani / 2005]

Practice prize report—an assortment wide decision-support
system for dynamic pricing and promotion planning in DIY
retailing [2007]

A proposal for Unified Service Pricing Framework [CSS /2009] –
web services stack

Already well established pricing techniques:
 Yield management
 Dynamic pricing

Inventory-based models
(stock, demand over time, customer types…)

Data driven models
(amazon, yahoo, airline industry)

Auction based models

Game theory models

Machine learning models
(single agent, multiple agents, reinforcement
algorithms, Markov decision process)

Simulation models

Our architecture proposal extends already known service pricing models (iRMU,
VTPES, SGEE)
see article in CSS209 - Collaboration, Software and Services in Information Society

It provides additional pricing information and simulation options (dynamics)

Simulation modules of the pricing system are related to the layers in webservices technologies stack
Service
billing
Workflow
management
module
Suggestion
module
Resource
pricing
Simulation module
(market simulations)
Traffic
measuring
Dynamic service pricing
modules
Service pricing
modules
Management & Security
Discount
module
Service flow
SLA server
module
1.
2.
3.
4.
5.
6.
7.
Service offerings are registered in the main repository by
service providers.
Client is selecting services for business process realization.
During service selection client is informed of similar
services with better offerings.
Semantic module is also monitoring service selection
process and suggests similar processes with special
offerings.
Client confirms service selection – a new usage contract is
defined.
Client confirms usage contract and enters additional
payment information. Usage contract defines payment
strategies and billing cycle.
Client can start using requested services.
Not all service offerings in the main repository are from the
same service provider
Services user selected for the realization of the business
process are contained in different SLA layers between
various service providers
Workflow:
1.
2.
3.
4.
5.
Service provider publishes service in the main repository.
Client can consume the service that is not composed in the business
process.
During publishing process service providers offers his starter pricing
offerings on the server. A template SLA for the provider is created.
Other providers negotiate and define their provider contracts based
upon provided template.
Service can be consumed in various business processes; the only
condition is the existence of intermediate SLA agreements.

Customer Relationship Management is at the
core of the dynamic pricing and offerings
(customers profile, interests, sites visited,
history-social engineering, google ad
words…)

DIY retail pricing decision support system
DoCoMo mobile platform


To develop a modular and step based process
for the development and support of dynamic
service pricing system. This includes
 Data mining
 Dynamic pricing
 Customer profile
 Market analysis/competition
 Region specifics





Define CRM system and categorization of it
for decision support and service offerings
Define the service system (organization
based or wider specter: problem CRM data)
Usage of existing prediction and estimation
methods for service price and service
proposal / semantics?
Data, data, data….?
Business process integration into
organization
Thank you….