OPERATIONS RESEARCH

INTRODUCTION TO OPERATIONS RESEARCH

Operations Research is a new branch of Mathematics dealing in the optimization problems in real-life situations. It is also a quantitative technique to deal many management problems, In this discipline ,we study cost minimization of various inventory problems, the minimization of transportation costs of sending goods from various warehouses to different centers, the profit maximization or cost minimization in linear programming models, the assignment of different person to different jobs so that total time taken to perform the jobs is minimized, the congestion problem in traffic places, airline counters, supermarket, to find out of the waiting time of customers in the queue, the project completion time with limited resources and many other similar problems.

ORIGIN OF OPERATIONS RESEARCH

The germination of the concept Operations Research occurred during World War 1. In England in the year 1915, F.W. Lanchester attempted to treat military operations quantitatively. He derived equations relating the outcome of a battle to both the relative numerical strength of the combatants and their relative manpower. He modeled a situation involving strategic choices and then tested that a model against a known situation. During the same period, Thomas Alva Edison in America was studying the process of Anti-submarine warfare. He devised a war game to be used for simulating problems of naval manoeuvre. In 1917, A.K. Erlang, a Danish Mathematician has developed solutions for some waiting line problems. In 1915, F.W.Harris had developed the first model on an inventory problem for economic
1

lot size In 1930 W.Leontieff developed a linear programming model representing the entire United status economy. Active research works were done during World War II in Great Britain and United States of America.

DEFINITION OF OPERATIONS RESEARCH

Operations Research was initially a subject dealing with military operations during the World War II. Later several techniques were developed to suit much of humanities progress in science, technologies, business administration, etc. A vast variety of fields is brought in the purview of this branch of science. There are served definitions for O.R. They only specify the applications of the discipline. None of the definitions is well defined. We mention some of the definitions here.

DEFINITION BY DCLARKE
Operations Research is defined as the art of winning wars without actually fighting.

DEFINITION BY ACOFF, ANNOFF AND CHURCHMANN

O.R. is the application of scientific methods, techniques and tools to problem involving the operations of system so as to provide those in control of the operations with optimum solution to the problem.

DEFINITION BY T.L.SASTRY
O.R. is the art of giving bad answers to problems where otherwise worse answers are given.

DEFINITION BY JAMES LUNTRY

O.R. is the sophisticated name given to multidisciplinary problemoriented approach to the top management problem. It involves the applications of scientific methods in situations where executives require description, prediction, and comparison for the purpose of decision making.

DEFINITION BY AMERICAN SOCIETY OF O.R.

O.R. is an experimental and applied science devoted to observe, understanding and predicting the behaviors of purposeful man-machine system, many of these definitions only broad line the applications of O.R. to war, industry, management and humanity progress.

APPLICATIONS OF OPERATIONS RESEARCH

Here we mention only some of the areas where O.R. techniques can be applied. This science is applied widely in areas of accounting facilities, planning, finance, manufacturing, marketing, purchasing and in organizations and government and quasi government activities. We mention some of the applications of O.R. in the above areas. Cash flow planning, credit policy planning of delinquent account strategy are some of the areas in accounting where O.R. techniques are used. Warehouses locations, Transportation loading and unloading, factory size and location, Hospitals planning are some of the areas of facilities planning . In finance, it is a applied to qualitative study of investment analysis, portfolio management, dividend policy, etc. In marketing, O.R. is applied to study the selection of product-mix, prediction scheduling time, advertising allocation, etc.
3

In order to arrive at an optimal solution to the problems in O.R., first we have to construct a model. Once a project is selected, we have to describe the problem as a model. The model should describe all the features of the problem. We have to express the description of the problem in a mathematical formulation. This formulation has not be done satisfying all the assumptions of the problem.

MODELS AND MODELLING

Modelling a real life situation helps us to study the different behavior of the problem corresponding to the description of the problem. Great efforts have been taken by experts to model business situations, military operations, motion of planets and stars, congestion in traffic places and so on. A model is an abstraction on an idealized representation of a real life problem. The object of a model is to provide means for analyzing the behavior of the system for future improvement. A map of multiple activity chart, a project network, the representation of the behavior of a queuing system, a model to forecast the future, based on the past and the present factors of a time series, etc.. are all examples of models, A model can be a picture, map, a curve or an equation. The reliability of the decision drawn from the model may depend upon the validity of the model on the basic assumption on which the model is built. Modelling is the essence of operations research building a model helps us to convert the complexities and uncertainties of a decision making problem into a concrete logical structure which is amenable to formed analysis. A model is a vehicle for arriving a well structured problem of reality. A commentator of a cricket match describes the play as a model to enable us to predict the future course of events of the play. It is a descriptive model available for further analysis. It is not always possible to analyse a situation only with the description of the
4

situation we have to formulate the problem into concrete mathematical representation in the form of a curve, graph or equations. Models, could be classified as iconic model, analogue models and symbolic model.

ADVANTAGES A iconic model is concrete. It is easy to construct the model. It is easy study the model then the system itself. DISADVANTAGES
This model is not suited for further manipulation. It cannot be used to study the changes in the operation of the system. It is not possible to make any modification of the model. Adjustment with changing situations cannot be done in this model.

ANALOGUE MODEL
In an analogue model, one set of properties is used to represent another set of properties. After analysing the model for decision making the results of the analysis can be re-interpreted in terms of the original system. For example, Contour lines on a map are analogues of elevation as they represent the size and fall of heights, Graphs are analogues as distance is used to represent a wide variety of variables such as time, percentage, weight, etc. It is easier to manipulate the analogue model. But it is less specific and less concrete.

SYMBOLIC MODEL
Symbolic models employ a set of mathematical symbols and functions to represent the decision variables and their functions to describe the behavior of the system. Almost all the models in O.R. are symbolic model.

ADVANTAGES
These models are most abstract and most general. These models are amicable for experimental manipulation. They yield reasonably good results to the real life problem. A good model should have the following characteristics : It should be capable of taking into account new formulations with having any significant change in its frame. The assumptions should be well defined and the number of assumptions should be as small as possible. The assumptions should be simple and coherent. Only a limited number of variables should be used. It should be acceptable to parametric treatment.

ADVANTAGE OF A MODEL
It is a description of a physical problem. It gives a systematic approach to a problem and is subject to logical treatment. It is easy to make decisions based on a model. If a model is built on a broad based assumption, it is easy to modify it according to new situations. Model help us finding avenues for new research and improvement in a system.

LIMITATION OF A MODEL
Models are only an attempt in describing a system and should be taken to be as absolute representation of a system. Model constructed is valid only if all the assumptions of the model are true in the system for which the model is constructed.

Validity of the model is subject to experimental testing.

PERT PROGRAM EVALUATION AND REVIEW TECHNIQUE

INTRODUCTION
Network scheduling is a technique used for planning and scheduling large projects in the fields of construction, maintenance, fabrication purchasing, computer system installation, research and development designs etc. The technique is a method of minimizing trouble spots, such as, production bottlenecks, delays and interruptions by determining critical factors and coordinating various parts of overall job. There are two basic planning and control techniques that utilize a network that to complete a pre determined project or scheduling. These are : Program Evaluation and Review Technique(PERT) and the Critical Path Method(CPM) several variations of these have also been developed. One such important variation being the Review Analysis of Multiple Projects(RAMP) which is useful for guiding the activities of several projects at one time.

NETWORK ANALYSIS
CPM was developed in 1957 by J.E. Kelly of Remington and M.R Walker of Dupont to aid in the scheduling of routine plant overhaul, maintenance and construction work. This method differentiates between planning and scheduling. Planning refers to the determination of activities that must be accomplished and the order in which such activities should be performed to achieve the objectives of the project.
8

PERT was developed in the late 1950s by the US Navy Special Projects Office in operation with the management consulting firm of Booz , Allen and Hamilton. The technique received substantial favourable publicity for its use in the engineering and development program of the Polaris missile, a complicated project that had 250 prime characters and over 9000 sub characters. But now this technique is very popular in the hands of project planner and controller of various departments in government and in industry. In PERT, we usually assume that the time to perform the activity is uncertain and as such three time estimates are used.

METHODOLOGY OF PERT/CPM NETWORKS

The methodology involved in applying PERT for any project may be split into the following steps: PROJECT PLANNING The purpose of this is to identify all important events/activities which are essential for completion as well as making up of the project and their dependence upon one another is shown explicitly in the form of a network. TIME ESTIMATION Estimates of the time required perform each of network activities are made, the estimates are based upon manpower and equipment availability

and certain assumptions that may have been made in planning the project. By incorporating the time required for completing each of the activities in the network, the project duration as well as the criticality of the activities are found. At this stage it is also possible to compute the probability of completing the project or a part of the project by a specified time.

 SCHEDULING The scheduling computations give the earliest and the latest allowable start and finish time for each activity, and as a by product, they identify the critical path through the network, and indicate the amount of slack time may associated with the non critical paths. TIME COST TRADE OFFS If the scheduled time to complete the project as determined in step 3 satisfactory, the project planning and scheduling may be complete However, if one interested in determining the cost of reducing the project completion time. Then time cost trade-offs of activity performs time must be considered for those activities on the critical and near critical path(s). RESOURCE ALLOCATION The feasibility of each schedule must be checked with respective manpower and equipment requirements. Establishing complete feasibility of a specific schedule may require replanning and rescheduling or time-cost trade-offs. Hence a final solution may require the performance of a number of cycles of steps 3, 4 and 5. PROJECT CONTROL When the network plan and the schedule have been developed to a satisfactory extent, they are repaired to final form for use in the field. The project is controlled by checking progress against the schedule, assigning and scheduling manpower and equipment, and analysing the effects of delays.

10

PROBABILTY CONSIDERATIONS IN PERT

The network methods discussed so far may be termed as deterministic, since estimated activity times are assumed to be the expected values. But no recognition is given to the fact that expected activity time is the mean of a distribution of possible values which could occur. Under the conditions of uncertainty, the estimate time for each activity are PERT network is represented by a probability distribution. This probability distribution of activity time is based upon three different time estimates mode for each activity. These are as follows: to = the optimistic time, is the shortest possible time to complete the activity if all goes well. tp = the pessimistic time, is the longest time that an activity could take if everything goes wrong. tm = the most likely time, is the estimate of the normal time an activity would take. If only one time where available, this would be it. Otherwise it is mode the probability distribution.

PROBABILITY OF MEETING THE SCHEDULE TIME

With PERT, it is possible to determine the probability of completing a contract on schedule. The scheduled dates are expressed as number of time units from the present time. Initially they may be the latest time, T L, for each event, but after a project is started we shall know how far it has progressed at any given date, and the scheduled time will be the latest time if the project is to be completed on its original schedule.

11

The probability distribution of times for completing an event can be approximated by the normal distribution due to the control limit theorem. Thus the probability of completing the project by schedule time(TS) is given by: Prob (z<(Ts-Te)/) the standard normal variate is given by, Z=(Ts-Te)/e Where Te = Expected completion time of the project. e = Number of the standard deviations the scheduled time lies from the expected time. (i.e) the standard deviations of the scheduled time. Using the commutative normal distribution table, the corresponding value of the standard normal variate is read off. This will give the require probability of completing the project on schedule time.

RULES OF NETWORK CONSTRUCTION

For the construction of a network, generally, the following rules are followed Each activity is represented by one and only one arrow. Each activity must be identified by its starting and end node which implies that.

i.

Two activities should not be identified by the same completion events, and
12

ii.

Activities must be represented by their symbols are by the corresponding ordered pair of starting completion events.

Notes are numbered to identify an activity uniquely Tail node should be lower than the head node, of an activity. Between any pair of nodes, there should be one and only one activity, however more than one activity may emanate from and terminate to a node. Arrows should be kept straight and not curved or bent.

NUMBERING THE EVENTS

After the network is drawn in a logical sequence, every event is assigned a number. The number sequenced must be such so as to reflect the flow of the network. In event numbering, the following rules should be observed: a. Event number should be unique. b. Event numbering should be carried out on a sequential basis from left to right. c. The initial event which has all outgoing arrows with no incoming arrow is numbered 0 or 1. d. The head of an arrow should always bear a number higher than the one assigned at the tail of the arrow. e. Gaps should be left in the sequence of event numbering to accommodate subsequent inclusion of activities, if necessary.

13

BASIC CONCEPTS OF NETWORK ANALYSIS

A fundamental ingredient in both PERT and CPM is the use of network system as a means of graphically depicting the current problems or proposed project. Because of its importance to a basic understanding of both PERT and CPM, the network concept will be examined. When a network is being constructed, certain conventions are followed to represent a project graphically, for it is essential that the relationship between activities and events are correctly depicted. Before illustrating the network representation, it is necessary to define some of the concepts.

ACTIVITY
All projects may be viewed as being composed of operations or tasks called activities, which require the expenditure of time and resources for the accomplishment. An activity as depicted by a single arrow ( ) on the project

network. The activity arrows are called arcs. The activity arrow is not scaled, the length of the activity time is only a matter of convenience and clarity, and does not represent important of time. The head of the arrow shows the sequence or flow of activities. An activity cannot begin until the completion of the preceding activities. It is important that activities be defined so that beginning and end of each activity can be identified clearly.

PREDECESSOR ACTIVITY
Activities that must be completed immediately prior to the start of another activity are called predecessor activities.

14

SUCCESSOR ACTIVITY
Activities that cannot be started until one or more of the other activities are completed, but immediately succeed them are called successor activities.

CONCURRENT ACTIVITY
Activities which can be accomplished concurrently are known as concurrent activities. It may be noted that an activity can be a predecessor or successor to an event or it may be concurrent with one or more of the other Activities.

EVENT
An Event represent a specific accomplishment in the project and takes place at a particular instant of time, and does not, therefore, consume time or resources. An event in a network is a time oriented reference point that signifies the end of the activity and the beginning of another. Events are usually represented in the project network by circles (o). The event circles are called nodes. Therefore, a major difference between activities and events is that activities represent the passage of time where as events are point in time. All activity arrows must begin and end with event nodes as shown below

Start event

Activity

Finish event

15

MERGE EVENT
Where more than one activity comes and joins, the event is known as merge event

BURST EVENT
When more than one activity leaves an event, the event is known as burst event.

MERGE AND BURST EVENT

An activity may be a merge and burst event simultaneously as with respect some activities it can be merge event and with respect to some other activities it may be burst event .

Merge event

Burst Event

Merge & Burst

DUMMY ACTIVITY
In most projects many activities can be performed concurrently or simultaneously. It is possible that two activities could be drawn by the same beginning and end events, In situations where two are more activities can be performed concurrently, the concept of dummy activity is introduced to solve this problem. Therefore there will be only one activity between two events.

16

An activity which does not consume either any resource or time is known as dummy activity. A dummy activity represented by dotted line in the network diagram.

Predecessor
Activity

Successor
Activity

Dummy activity

PERT SYSTEM OF THREE TIME ESTIMATE

The traditional single estimate of duration of any activity is replaced by three time estimates in PERT system an optimistic, a pessimistic, and a most likely time.

OPTIMISTIC TIME (a or to)

The time estimate of an activity when everything is assumed to go well as per plan. In other words, it is the estimate of the minimum possible time, which an activity takes in completion under ideal conditions. However no provisions are made for breakdown, delays, etc

MOST LIKELY TIME (m or tm)

The time which the activity will take most frequently performed a number of times the model value.

PESSIMISTIC TIME (tp)

The unlikely but possible performance time if whatever could go wrong, goes wrong in series. In other words it is the longest time the activity can
17

conceivably take. This however does not include major catastrophies like labour strikes, acts of God unrest, etc.

EST- It means Earliest start time for an activity represent the time at which an
activity begins at the earliest.

EFT- EFT means Earliest finish time of an activity is it earliest start time +
(plus) the required time to perform the activity.

LFT- LFT means latest finish time. Latest finish time of an activity represent the
latest by which an activity must be completed without delaying the completion of project.

LST- Latest start time for an activity is the Latest finish time -(minus) the
activity duration methods.

FORWARD PASS METHOD (For Earliest Event Time)

Based on fixed occurrence time of the initial network event, the forward pass computation yields the earliest start and earliest finish times for each activity and indirectly the earliest expected occurrence time for each event.

BACKWORD PASS METHOD (For latest allowable time)

The latest occurrence event time (L) specifies the time by which all activities entering into that event must completed, without delaying the total

18

project. These are computed by reversing the method of calculation used for earliest event times.

CRITICAL PATH METHOD

The longest path is called the critical path. An activity is said be critical if the delay in its start will delay the project completion time.

PERT-ALGORITHM
The various step involved in developing PERT network for analyzing any project are summarized below Make a list of activities that make up the project including immediate Predecessors. Making use of step1 sketch the required network. Denote the most likely time by tm, the optimistic time to and pessimistic time by tp. Using beta distribution for each activity duration the expected time t e for te = (to+tm+tp)/6 Tabulated various time (i.e) expected activity times, earliest and latest times and mark the EST and LFT on the arrow diagram. Determine the total float for each activity by taking the difference between EST & LFT. Identify the critical activities and connect them with the beginning node and the ending node in the network diagram by double line arrows. The critical path and expected date of completion of the project Using the values of tp and to to compute the variance (2) of each activity . This is done with the following formula, 2

= [(tp-to)/6]2
19

 Compute the standard normal deviate Due date Expected date of completion

Zo

project variance
Use standard normal tables to find the probability p(z zo) of completing the project within the scheduled time, where Z ~ N(0,1).

20