Accident Prevention Using Case Based Reasoning in Construction Industry

You might also like

Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 38

PROJECT REPORT

PRESENTATION

ACCIDENT PREVENTION USING


CASE BASED REASONING IN
CONSTRUCTION INDUSTRY

Presenter
Ch. Mahesh Kumar
M.E (CEM)
Roll No: 0109 -1177

Supervisor
Dr.V.S.S KUMAR
Outline
• Objectives
• Need for CBR
• About CBR
• Advantages & Disadvantages of CBR
• Different Approaches of CBR
• Methodology
• Cases / Case Study
• CBR Tools
• Soft Computing Approach
2
Objectives
 The primary objective of the project thesis is
intended to point out the need for safety in
construction industry and to bring forth the role of
CBR in accident prevention in construction projects.

 To identify the set of attributes and to store the data


in the form of a case, which makes the process of
decision making simple.

 To find out the similarity value and based on this


pick the most relevant / suitable to the source case.

 To explore the different CBR Tools.


3
Need for Safety
S.No Cause/Reason % of Fatal Accidents

1 Falling of Persons 45

2 Falling of Material 14

3 Transport 14

4 Lifting Equipment 7

5 Excavation 7

6 Electricity 6

7 Other Causes 7

4
Major areas of Risk
Prevention of accidents is the major aim of safety management.

Accidents are likely to occur causing physical injury, causalities and


loss of money. In order to prevent accidents at construction sites,
certain safety measures need to be taken in the following major
activities prone areas.

• Excavation
• Drilling & Blasting
• Road works (Hot Bituminous)
• Scaffolding, Ladders, Formwork equipment
• Fabrication & Erection
• Storage
• Demolition 5
Types of Hazards at Construction Sites
• House keeping
• Means of access & egress
• Lightning and ventilation
• Dangerous and harmful environment
• Fume/Gases due to welding and cutting
• Dust, gases,& fumes
• Excessive noise
• Corrosive substances
• Lifting and carrying of excessive weight
• Fall of person, objects and materials
• Entry of unauthorized persons
• Slipping, tripping, cutting, drawing and falling hazards
• Safety in the use of Electricity.
• safety in use and handling of explosives
• Storage of explosive and blasting agents
• Safety in the erection, use and dismantling of scaffolds
• Fire Hazards 6
Need for CBR
A case-based reasoning solves new problems by using or adapting
solutions that were used to solve old problems. In simple case-based
reasoning means learning from previous experiences.

A case based decision support system helps to exploit data so that


smarter decisions can be made in less time and / or at lower cost. It is a
methodology that models both human reasoning and thinking.

The Basic Principle in this decision support system is that once data
pertaining to a case is fed to the system, the decision support system
compares these details with those for a wide range of cases stored in
memory; it then retrieves similar cases from memory.

7
About CBR

Case-based reasoning (CBR) is a rapidly


emerging AI technology that can use past
experiences (cases) to solve current problems.

The idea of reasoning from relevant past cases


is appealing because it acts to the process an
expert uses to solve "new" problems quickly
and accurately.

8
Advantages of CBR

• Can use when have “weak” domain

• Reduces knowledge acquisition effort

• Can propose solutions from incomplete problem statements

• Improves over time as case base grows

• High user acceptance

9
Advantages of CBR Decision Supporting System

• Finding the solution to complex problems quickly

• Discovering decision knowledge hidden in data

• Transferring experience from skilled specialists to


users

• Building a database by sharing individual experiences.

10
Working Methodology of CBR Cycle:
The 4 - R approach

RETRIEVE RETAIN
find similar integrate in
problems case-base

CBR
REUSE REVISE
propose solutions adapt and repair
from retrieved cases proposed solution
11
CBR Solving Problems

Solution
Retain Review

Database Adapt

Retrieve
Similar
New
Problem
12
CBR Process

1. Search for and retrieve the most relevant cases from case memory

2. Reuse the retrieved cases to solve the current problem

3. If necessary, adapt the cases to better meet the current problem

4. Retain any newly created cases in memory for future use.

13
Different approaches of CBR

1. Textual Approach

2. Conversational Approach

3. Structural Approach

14
Textual Approach
Cases are recorded as free text. Cases
may be product descriptions, service
reports etc.

The CBR retrieval engine uses various


keyword matching techniques to
retrieve cases.

15
Conversational Approach

This approach to case based reasoning


involves describing the cases by developing
groups of questions, which are arranged into
a hierarchy to correspond to the hierarchy of
cases, from the general to the specific.

This approach resembles the construction of


a rule base of an expert system using
decision trees.

16
Structural Approach
A common structure for knowledge
representation called the domain model is first
defined.

This domain model specifies a set of attributes


used to represent a case and also structures the
set of attributes.

Each attribute must be a field in a database and


each value must be standardized.

17
The 4-RE’s
Retrieving
Retrieving a case starts with a problem description and ends when a best matching
case has been found. The subtasks involve:
1. Identifying a set of relevant problem descriptors;
2. Matching the case and returning a set of sufficiently similar cases
3. Selecting the best case from the set of cases returned
2. Reusing
Reusing the retrieved case solution in the context of the new case focuses on
identifying the differences between the retrieved and the current case. Also identify
the part of a retrieved case which can be transferred to the new case. Generally the
solution of the retrieved case is transferred to the new case directly as its solution
case.
3. Revising
Revising the case solution generated by the reuse process is necessary when the
solution proves incorrect. This provides an opportunity to learn from failure.
Retaining
Retaining the case is the process of incorporating whatever is useful from the new
case into the case library. This involves deciding what information to retain and in
what form to retain it; how to index the case for future retrieval; and integrating the
new case into the case library.
18
CBR Tools

• ART*Enterprise
• Case-1
• CasePower
• CBR2
• Eclipse
• ESTEEM
• KATE
• ReCall

19
20
METHODOLOGY ADOPTED
The first step is to input the 20 cases that constitute the case
base.

Building the case base involved not only inputting data for
every input feature for each of the 20 cases but also
adopting a method of similarity assessment.

A CBR system derives its power from its ability to retrieve


relevant cases quickly and accurately from its memory.

Indexing is the process of comparing the target case with


cases stored in the case base using an appropriate type of
similarity assessment and an appropriate type of feature
matching to produce a similarity score. 21
Types of Case Representation
• Flat feature-value list
• Object Oriented representation
• Graph representation

• The choice of representation is


– Dependent on requirements of domain and task
– Structure of already available case data

22
CASE REPRESENTATION
Case_id: 1

Problem features

Location :
Accident Type :
Company Name :
Construction Type :
Cost :
Accident Description :

Solution

Solution adopted :

23
SIMILARITY ASSESSMENT

The most commonly used techniques that can be applied in the


retrieval phase of a CBR system can be

Nearest Neighbor Techniques


Induction Algorithms
Fuzzy Logic
Database Techniques such as SQL

Similarities are usually normalized to fall with in the range 0 to 1, where 1 means a
perfect match and 0 indicates a total mismatch.

24
Nearest Neighbour Techniques
The similarity of the problem (target) case to a case in the case-library for each
case attribute is determined.

This measure is multiplied by a weighting factor. Then the sum of the similarity
of all attributes is calculated to provide a measure of the similarity of that case in
the library to the target case. This can be represented by the equation:

where:

T is the target case


S is the source case
n is the number of attributes in each case
i is an individual attribute from 1 to n
f is a similarity function for attribute i in cases T and S and
w is the importance weighting of attribute i

This calculation is repeated for every case in the case-library to rank cases by similarity to the target.
Similarities are usually normalised to fall within a range of 0 to 1 (where 0 is totally dissimilar and 1
is an exact match) or as a percentage similarity where one hundred percent is an exact match.

25
Nearest Neighbour Retrieval
• Retrieve most similar
• k-nearest neighbour
– k-NN
– like scoring in bowls or curling
• Example

– 1-NN
– 5-NN

26
Sample Case

27
Technical
Diagnosis of Car
Faults

28
Problem to be solved

29
How CBR solves problems
• New problem can be solved by
– retrieving similar problems
– adapting retrieved solutions
• Similar problems have similar solutions

P? P P PP
P P
P P

S S S
S S X S S
S S
30
31
Similarity Computation for Case 1

Figure Source: R. Bergmann, University of Kaiserslautern


32
Similarity Computation for Case 2

Figure Source: R. Bergmann, University of Kaiserslautern


33
Similarity Measurement
• Purpose: To select the most relevant case
• Basic Assumption: Similar problems have similar solutions
• Similarity value between 0 and 1 are assigned for feature
value pairs
• E.g.: Feature: Problem
Front Light does not work
Front Light does not work
Break Light does not work .4
.8
Engine doesn’t start

34
Similarity Measurement
• Feature: Battery Voltage .1
.9
12.6 13.6 12.6 6.7

• Different features have different importance


• Two kinds of Similarity Measures
– Local Similarity – similarity on feature level
– Global Similarity - similarity on case or object
level
35
Applications

• Failure prediction • Maintenance scheduling


– ultrasonic NDT of rails – Boeing 737 engines
for Dutch railways – TGV trains for SNCF
– water in oil wells for
Schlumberger
• Planning
• Failure analysis
– mission planning for
– Mercedes cars for
US navy
DaimlerChrysler
– route planning for
– semiconductors at
DaimlerChrysler cars
National Semiconductor

36
More Applications
• e-Commerce • Re-Design
– sales support for standard – gas taps for Copreci
products
– sales support for
customised products
• Personalisation • Formulation (recipes)
– TV listings from – rubber for racing tyres
Changing Worlds for Pirelli
– music on demand from
Kirch Media – tablets for AstraZeneca
– news stories via car radios
for DaimlerBenz 37
Thank you

38

You might also like