Pattern Model Stratification to Decision-making Support

of Transport Infrastructure Management

Tatyana Mikheeva

1 a

, Sergey Mikheev

1 b

and Oleg Golovnin

2 c

Department of Transportation Management and Control, Samara National Research University, Samara, Russia

Department of Information Systems and Technologies, Samara National Research University, Samara, Russia

Keywords: Pattern, Stratification, Decision-making Support, Transport Infrastructure.

Abstract: Design and engineering of patterns requires tools to enable these processes by means of universal creation

and dynamic modification of objects. The article introduces the basic definitions of pattern design (hereinafter

P-models), specifies the types and structures of patterns, determines the organization of object design based

on P-models. The introduced axiomatic and systematization are based on the use of original definitions and

form the concept of stratified pattern construction of objects and associations. A stratified pattern model is

considered as an information, methodological and implementation basis for the design of the decision support

system of transport infrastructure management. The synthesis of the decision support system of transport

infrastructure management is made on the fundamental methodological basis, determined by the object-

oriented paradigm. The synthesis of universal structures based on patterns is considered on the formal

declaration of objects invariant to their subject orientation. Along with the synthesis of methodical P-models,

the patterns of formalization and systematization of relevant information-logical and functional aspects of

transport infrastructure are described. The implementation P-models contains the effectiveness of the pattern

design of the decision support system of transport infrastructure management of the processes of modelling,

management, experimental studies.

1 INTRODUCTION

The system approach, as a scientific direction of

solving management problems, is the study of objects

and systems with the analysis of static structures and

dynamic functionals. The system is aggregated by a

set of components that are in a relationship, having

connections between themselves, forming integrity

and unity (Kabir et al., 2018). A complex

organizational and technical decision support system

for transport infrastructure management is a system

that emphasizes the cognitive and adaptive aspects of

its operation, which has the potential for self-

organization and development (Šelih et al., 2008;

Speranza, 2018). Attempts to formalize decision-

making support tasks have led to the emergence of a

number of approaches, methods, and intelligent

systems that help the decision maker in his work.

In all engineering disciplines, object-oriented

https://orcid.org/0000-0002-8485-8041

https://orcid.org/0000-0001-8489-2592

https://orcid.org/0000-0002-1418-2226

analysis, design, and programming, based on the

principles of decomposition, abstraction, and

hierarchy, are widely used (Coad and Yourdon,

1991). Modern methodologies of structural analysis

and design in CASE-tools, various knowledge

engineering technologies in general do not offer a

systematic procedure or formalism allowing to

“deduce” the structure of concepts and relations of the

subject area from the data available about it

(Golovnin and Mikheeva, 2018).

2 PATTERN STRATIFICATION

The methodology of formalizing, analysing, and

synthesizing a decision support system of transport

infrastructure management is based on an integrated

strategy for increasing the level of abstraction of the

models used, covering the realizable combinations of

498

Mikheeva, T., Mikheev, S. and Golovnin, O.

Pattern Model Stratiﬁcation to Decision-making Support of Transport Infrastructure Management.

DOI: 10.5220/0007729904980504

In Proceedings of the 5th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2019), pages 498-504

ISBN: 978-989-758-374-2

features of domain objects. The methodology is based

on the concept of object-oriented analysis and design

as a set of entities of the domain (objects) interacting

with each other, considering objects as instances of

certain classes that form the taxonomic hierarchy (Cai

et al, 2018).

Implementing projects to develop decision

support systems and business process modelling

creates situations where problem solving in various

projects has similar structural features. In many

object-oriented systems, you can find patterns

consisting of classes and interacting objects, with the

help of which you can solve specific design problems

that exist simultaneously in several systems. The

generalization and classification of such tasks and the

most successful ways to solve them led to the

emergence of patterns. The pattern approach is

characterized as any mathematical theory is a set of

special cases that are interpreted from a unified

position (Grenander, 1993). On the basis of object-

oriented design, the pattern from an abstract category

has become an integral attribute of modern CASE-

tools (Limayem, 2004; Maksimov et al., 2018).

The essential features of the intelligent transport

geographic information system for decision-making,

as an informational pattern model, are:

 the complexity and scale of the models that fill

the transport infrastructure management system,

expressed in a large number of types, using

alternative mechanisms of multiple inheritance and

polymorphic redefinition of object types properties,

using nested aggregate and selective constructions

and associations (Petzold and Freund, 1990; Vanier,

2004);

 the need to support queries to data in declarative,

predicative and navigation styles, the effective

implementation of basic manipulating operations

(Tomingas et al., 2015);

 wide context of using models in applications that

operate with data from one multidisciplinary

information scheme, as well as data from several

independent schemes (Kleppmann, 2017);

 interaction with external systems,

heterogeneous, hybrid, multifunctional with the

definition of the mechanism of interaction,

integration into a single space of data and functional

(Wood et al., 2015).

Definition 1. A pattern is a sample, a template

model, a formalized description of a frequently

encountered design problem; an effective in a given

context typical solution of a design (software)

problem. We define a pattern model as a P-model.

In modern software engineering, the creation of

software products relies on the explication of the

intellectual role of software and hardware concepts.

In this sense, the architecture of the transport

infrastructure management system should include

structural and functional components that,

individually or in certain combinations, are designed

to display the intelligent units of the overall design

scheme and the technological components with which

the intelligent units in question are generated and

interconnected.

According to the dominant cognitive paradigm for

abstracting any subject area, the categories of

“objects” and relations between objects are

recognized as primary and atomic.

The declaration of generalized properties inherent

in objects of the domain, and the generalization of

these properties to ensure the transition to the highest

level of the abstraction hierarchy, are inherent in

taxonomic object-oriented models. This abstraction

provides a modular (patterned) system design due to

a stratified rise from one level of the generalization

hierarchy to another.

It is reasonable to use object-oriented taxonomic

models as a methodological and informational basis

for designing a decision support system for managing

transport infrastructure. Unlike network, relational,

and hierarchical models, which declare only static

relationships, taxonomic models not only structure

information, but also ensure the “inheritance” of both

the class objects themselves and the processing

methods of object-oriented structures; the entire

inheritance hierarchy, which allows to synthesize

patterns.

Thus, taxonomy with stratification functional is

most relevant to the tasks of developing a transport

infrastructure management system.

Definition 2. The stratified P-model declares the

static regulation of the structured construction of

classes of objects and interclass relations.

The stratified P-model declares strata as objects

that receive ontological meaning and describe

elements of a nonlinear-dynamic system, which is the

decision support system of transport infrastructure

management. Interclass relations specified by the P-

model cannot be changed in the dynamics of the

simulation process. Static character of the regulations

does not mean the static nature of the objects

constructed under this regulation, they can be

dynamically reconstructed in accordance with the

structure of interclass relations in the model.

Visually, the P-model is represented by a set of

nested spherical strata defined by a triad

, ( , , ..., ),

S    

, where

is the

set of strata of the domain space,

, , ...,

  

Pattern Model Stratiﬁcation to Decision-making Support of Transport Infrastructure Management

499



Figure 1: Stratified P-model.

is the set of types of relations (connections) between

the strata of the space,



is the set of mappings. In

addition to traditional spatial, temporal and causal

relations, the system also contains specialized

relations characteristic of the subject area: set-

theoretic, logical, functional, topological,

hierarchical.

The spherical stratum is a class of objects and can

be decomposed to obtain the necessary level of detail.

Decomposition is represented by clipping a cone from

the common domain space (Figure 1). We define

mappings from any stratum to each, which allows us

to link data from different strata. At the same time, it

is necessary to ensure the preservation of strata

invariants — the set of essential and invariable

properties of the objects of each strata.

The stratified P-model of the decision-making

support system of transport infrastructure

management is defined as the tetrad of strata:

 

( ) , , ,P S A D M F

The central stratum of space is filled with basic

atomic patterns that have the indivisibility of their

semantics and functionality. In accordance with the

stratification methodology, the following strata are

defined as: declarative stratum, methodical stratum

and implementing (functional) stratum.

The declarative stratum is stratified into strata of

information about the objects of the subject area, their

immanent properties and architectural structures. The

knowledge representation stratum contains statically

declared data models and output algorithms that

operate on data structures and have a functional

independent of the content.

Methodical stratum is stratified into strata of

methods for collecting, processing, storing,

transmitting information about domain objects,

mathematical methods of intellectual analysis,

design, modelling, visualization, prediction of

situations arising.

The implementation stratum is stratified into the

strata of the data processing software of the

declarative stratum through the program

implementation of the methodical stratum methods

based on functional zoning patterns.

Stratum mapping is defined as:

AD

DM

DF

MF

The synthesis of universal structures based on

patterns is based on the formal declaration of objects

that is invariant to their subject orientation. Such an

approach to the development of a complexly

organized decision support system is provided by the

pattern representation of any object of transport

infrastructure in the form of an architecture consisting

of two parts: the formal (architectural), interaction

with which is carried out only through universal

declarative patterns, and the content filled with the

values of the intrinsic domain objects.

Along with the synthesis of methodical P-models,

patterns of formalization and systematization of

relevant information, logical and functional aspects of

transport infrastructure are defined, patterns of formal

transformation of initial information into pattern

specification that implements all key forms of

semantic abstraction, modelling, zonal management

and analysis of transport infrastructure based on

artificial intelligence.

VEHITS 2019 - 5th International Conference on Vehicle Technology and Intelligent Transport Systems

500

Implementing P-models contain the effectiveness

of the pattern design of the decision support system

of transport infrastructure management of the

modelling, management, and experimental research

processes. This gives a qualitative explanation of the

main phenomenon of the conceptualization phase in

the tasks of constructing complexly organized

systems — the possibility of the formation and

coexistence of various types of models within one

subject area.

Synergetic P-models act as part of a system

analysis of the general principles of the complex

system, which is considered as a decision support

system for the chaotic management of transport

infrastructure with the synthesis of self-organization

and self-development processes, taking into account

the restrictions imposed. The P-Integration synergetic

pattern is an architectural pattern for synthesizing a

decision support system that describes plug-in

interaction strata. The generalized pattern is the basis

of the design and the system as a whole, and

individual plug-ins that implement a narrow range of

specialized tasks of the domain. The P-Integration

synergetic pattern describes the limitations and

interfaces of plug-ins that make up the decision

support system, ensuring their integration and joint

operation. The integrating environment contains a

platform for executing transaction scenarios, basic

functionality for application interaction, logging

services and monitoring of the state of the integrating

environment.

The declarative, methodical and implementing

strata form the integrating environment of the P-

Integration synergetic pattern, which contains the

process of transferring data from the declarative

stratum, using methods from the methodical stratum,

assigning data and methods to the implementing

stratum.

Different patterns represent the internal affiliation

of the system, largely reflecting the peculiarities of its

specific implementation in solving problems of

transport infrastructure management, and all other

components are instances of various architectural

components.

The methodical stratum contains the patterns of

the supervisor of the decision support system, the

patterns of the masters of database design, the models

of objects and processes of the domain, the patterns

of the masters of the design of plug-ins. The Pattern

Database Wizard generates an object model base

template in accordance with the model it is specified.

In the generated class pattern of the model, proper

storage structures are created only in relation to those

classes whose objects will actually constitute object

models of the domain. Objects of classes describing

abstract, generalizing concepts may not be included

in object models. However, the properties and

methods described in these classes are represented in

an object-oriented database. This occurs in cases

where at least one class derived from “abstract”

classes has a status indicating the presence of objects

of this subclass in the object model. Model Wizard

automates the development of model supervisors. A

model supervisor is an architectural component of a

system through which all components of an

architecture form links. This unification allows us to

state a high degree of openness in the architecture of

the decision support system in managing the transport

infrastructure, its ability, as a tool, to evolve.

3 P-MODEL IMPLEMENTATION

We consider the implementation of the P-model of

the ITSGIS intelligent transport geo-information

system: the “Strategy_ITSGIS” pattern (Mikheev et

al., 2012).

Task. It is necessary to design changeable, but

reliable algorithms or strategies of behaviour.

Decision. A pattern encapsulates an object’s

algorithm, simplifying its specification and making it

independent of other algorithms and clients that use

it. Encapsulation strategy in a separate class allows

you to change the behaviour of the object

dynamically. “Strategy_ITSGIS” allows you to

configure a class by specifying one of the possible

behaviours. The mechanism of inheritance makes it

possible to isolate the functionality common to all

algorithms, thereby supporting a variety of algorithms

or behaviours. Obviously, to add a new strategy, you

need to create an independent class and implement

the necessary functions. Similarly, only one class will

be affected to change or delete a strategy. When

describing the behaviour of an object by several

methods, the “Strategy_ITSGIS” pattern reduces the

number of conditional statements in the program

code. The conditional operator will need no more than

once to select a specific strategy. You can use the

“Strategy_ITSGIS” pattern, for example, when

choosing an encryption algorithm.

Example A. A system for automatically

constructing a transport network model on a thematic

layer of an electronic map. When constructing a

model of a transport network consisting of several

“universal” sections, algorithms of the same meaning

are used, but used for different conditions of

dislocation of these sections on the map in ITSGIS.

These are algorithms for constructing intersections,

Pattern Model Stratiﬁcation to Decision-making Support of Transport Infrastructure Management

501

Figure 2: Synthesis of universal transport network structures in ITSGIS.

+ModelBuild()

MIA_Algorithm

+ModelBuild()

«interface»

IAlgorithm

+ModelBuild()

COMBI_Algorithm

+ModelBuild()

TMNN_Algorithm

+ButtonClick()

Level_Availability

IAlgorithm Algorithm;

Algorithm = DataAnalyzer.CreateAlgorithm();

Algorithm.ModelBuild();

Figure 3: “Strategy_ITSGIS” pattern.

pedestrian crossings, railway crossings, tunnels,

overpasses, bridges, hauls of various configurations

(dead ends, turns, broadenings, etc.). The site

construction object applies the appropriate algorithm

depending on the conditions of its construction and

visualization on the map (Figure 2).

Example B. The plug-in of the intellectual analysis

of the state of spatially coordinated objects. The

design of ITSGIS takes into account the extensibility

of the system in terms of adding new data research

algorithms, the use of which should not differ from

the use of existing ones in ITSGIS. The plug-in of the

intellectual analysis of the state of geo-objects allows

to determine the dependence of the level of accidents

on a part of the transport network on various factors

that are constantly present or occur during the

functioning of the transport infrastructure: the

presence of children’s educational institutions,

pedestrian crossings, road signs, emergency recovery

works, traffic accidents and other. Figure 3 shows the

class diagram for applying the “Strategy_ITSGIS”

pattern. The “Level_Availability” client class has

access to the interface assembly and therefore can

define a variable of type “IAlgorithm”. The client

class receives a link to one of the algorithms located

on the server through the creator class. The decision

about which class of algorithm to create is made by

the user, calling one of the overloaded methods of the

creator class. Adding a new class of algorithm to the

existing ones will not break the overall system

architecture. The “Strategy_ITSGIS” pattern

provides the extensibility of the ITSGIS system by

adding new algorithms for constructing a regression

model. Each algorithm is described in a separate class

VEHITS 2019 - 5th International Conference on Vehicle Technology and Intelligent Transport Systems

502

+ModelBuild()

PAdapter

IAlgorithm

+GetLinkTable()

+GetResult()

+Fill()

Map_Adapter

-SourceTable

-ResultTable

-IsSelected

Thematic_Layer

+GetLinkTable()

+GetResult()

+Fill()

PMap_Layer

IInitial_Data

+GetLinkTable()

+GetResult()

+Fill()

Data_Structure

Data_Table

Figure 4: “PMap_Layer_Adapter” pattern.

(“TMNN_Algorithm”, “MIA_Algorithm”,

“COMBI_Algorithm”) implemented by the

“IAlgorithm” interface. In the “Level_Availability”

client class, a variable of “IAlgorithm” type is

defined, which is initialized by an instance of one of

the classes of the algorithm chosen by the user.

Task. It is necessary to design a decision support

system for the chaotic management of the transport

infrastructure with the synthesis of self-organization

and self-development processes, taking into account

the restrictions imposed.

Decision. The “PMap_Layer_Adapter” pattern of

synthesis of the thematic layer of the electronic map

of the geographic information component of the

intelligent transport infrastructure management

system is based on the “PAdapter” pattern for

converting information from a map display of objects

into a form for the functioning of neural network

artificial intelligence methods. Geographical

visualization of information is presented on an

electronic map in the form of thematic layers, each of

which displays a specific type of geospatial attribute-

oriented models of geo-objects.

Example A. Plug-in analysis of correlated spatial-

coordinated geodata based on the processes of self-

organization and self-development of various types of

spatial geo-objects. Methods for analysing correlated

spatially coordinated geo-data are developed on the

basis of implementation patterns.

The “PMap_Layer_Adapter” pattern converts the

interface of one class to the interface of another,

aggregates the spatially coordinated data rendered on

the corresponding thematic layers and stored in the

corresponding system database tables, and operates

with artificial intelligence methods that return data in

general form data structures.

The class diagram of “PMap_Layer_Adapter”

pattern for the system of intellectual analysis of

spatial-coordinated geo-objects is presented in Figure

4. “IAlgorithm” is an interface that is inherited by the

algorithms for studying the attributes of transport

infrastructure objects. “Map_Adapter” is an adapter

that brings the interface of spatially coordinated

objects stored in instances of “Data_Table” and

“Thematic_Layer” to the “IAlgorithm” interface.

4 CONCLUSIONS

The stratified pattern model is used as an

informational, methodological, and implementation

basis for designing a decision support system of

transport infrastructure management.

The synthesis of a decision support system of

transport infrastructure management is based on a

fundamental methodological basis, defined by an

object-oriented paradigm that establishes the primacy

and atomicity of object categories, taxonomic and

aggregate patterns of properties and relations between

objects.

REFERENCES

Cai, S., Gallina, B., Nyström, D., Seceleanu, C., Larsson,

A., 2018. Tool-supported design of data aggregation

processes in cloud monitoring systems. In Journal of

Ambient Intelligence and Humanized Computing.

Coad, P., Yourdon, E., 1991. Object-oriented analysis,

Englewood Cliffs, NJ: Yourdon press.

Golovnin, O. K., Mikheeva, T. I., 2018. Attribute-driven

network-centric urban transport process control system

modeling. In Journal of Physics: Conference Series.

Grenander, U., 1993. General Pattern Theory, Oxford

University Press.

Kabir, S., Yazdi, M., Aizpurua, J. I., Papadopoulos, Y.,

2018. Uncertainty-aware dynamic reliability analysis

framework for complex systems. In IEEE Access.

Pattern Model Stratiﬁcation to Decision-making Support of Transport Infrastructure Management

503

Kleppmann, M., 2017. Designing data-intensive

applications: The big ideas behind reliable, scalable,

and maintainable systems, O'Reilly Media.

Limayem, M., Khalifa, M., Chin, W. W., 2004. CASE tools

usage and impact on system development performance.

In Journal of Organizational Computing and

Electronic Commerce.

Maksimov, M., Fung, N. L., Kokaly, S., Chechik, M., 2018.

Two decades of assurance case tools: a survey. In

International Conference on Computer Safety,

Reliability, and Security. Springer.

Mikheev, S. V., Mikheeva, T. I., Malykova, K. A.,

Mikhaylov, D. A., 2012. Disposition of objects of a

transport infrastructure in the GIS ITSGIS. In

CSIT’2012, 14th International Workshop on Computer

Science and Information Technologies.

Petzold, R.G., Freund, D.M., 1990. Potential for geographic

information systems in transportation planning and

highway infrastructure management. In Transportation

Research Record.

Šelih, J., Kne, A., Srdić, A., Žura, M., 2008. Multiple‐

criteria decision support system in highway

infrastructure management. In Transport.

Speranza, M. G., 2018. Trends in transportation and

logistics. In European Journal of Operational

Research.

Tomingas, K., Kliimask, M., Tammet, T., 2015. Data

integration patterns for data warehouse automation. In

New Trends in Database and Information Systems.

Springer.

Vanier, D. J., 2004. Geographic information systems (GIS)

as an integrated decision support tool for municipal

infrastructure asset management. In Proceedings of the

CIB 2004 Triennial Congress.

Wood, T., Ramakrishnan, K. K., Hwang, J., Liu, G., Zhang,

W., 2015. Toward a software-based network:

integrating software defined networking and network

function virtualization. In IEEE Network.

VEHITS 2019 - 5th International Conference on Vehicle Technology and Intelligent Transport Systems

504