FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003
COMPLETE SPECIFICATION
(See Section 10 and Rule 13)
Title of invention:
SYSTEM AND METHOD TO PROVIDE PREDICTIVE ANALYSIS TOWARDS PERFORMANCE OF TARGET OBJECTS ASSOCIATED WITH ORGANIZATION
Applicant
Tata Consultancy Services Limited
A Company Incorporated in India under The Companies Act, 1956
Having address:
Nirmal Building, 9th Floor,
Nariman Point, Mumbai 400021,
Maharashtra, India
The following specification particularly describes the invention and the manner in which it is to be performed.

FIELD OF THE INVENTION
[001] The present invention in general relates to a field of performance analysis. More particularly, the present invention relates to a system and method to provide predictive analysis of performance of one or more target objects associated with an organization with respect to one or more parameters.
BACKGROUND OF THE INVENTION
[002] In recent years, organizations are increasingly focused on monitoring processes, their performance and their evaluation. Organizations are attempting to manage their performance by tracking and measuring it across dimensions. Organization performance is measured in terms of effectiveness in achieving their goals towards meeting targets that are aligned with some objectives associated with the target. Majorly, performance is measured in terms of ability to effectively deploy services to a client to further manage a satisfaction level. Client satisfaction is of utmost important and a critical component of any service industry.
[003] Meeting the target performance has become an essential factor at a strategic level and an important part of operational excellence. It has become a pre-requisite to improve customer loyalty by enabling customer retention to ensure repeat service retrieval and increase share of wallet. This may be achieved by meeting the client requirements or sometimes performing beyond expectations of the requirements. Critical research has been carried out in the past, to provide various methods to measure and improve organization performance to attain targeted client satisfaction.
[004] US2012059686 discloses a method to provide recommendation related to key parameters having the highest probability to increase the primary performance. The primary performance indicator comprises an ordinal data point having a calculated ordinal data level. A set of key drivers are determined as having an influence on the primary performance indicator. The method more specifically includes calculation of probability to increase in primary performance indicator based on increase of each member of a set of key drivers by a single ordinal data level. However, current method is

associated with vendors providing IT services to external customers and considering factors. The considered factors are more relevant from delivery and service management perspective. Also, the method is more particularly used for customer management.
[005] Another method disclosed in KR20100107851 provides online client satisfaction survey and a statistical analysis to supply a product or service innovation opportunity by calculating a (Customer Satisfaction Index) CSI for a product and service in real time. Here, a CSI is calculated by performing a customer satisfaction survey in real time by utilizing each kind of database pool. However, this method mainly focuses on real time building of a questionnaire database when customer chooses for a survey, it pulls up the right questions from the database and when right questions are not present in the database then the new right questions are added to the database. Thus, the method has limited approach of calculation of CSI and does not teach improving the organization performance.
[006] US8311874 provides a method for customer relationship evaluation and resource allocation. The method teaches evaluation and enhancement of customer relationship by generating actionable inferences through the measurement and analysis of both customer satisfaction and customer importance. A framework is provided to flag critical relationships for additional management focus and review.
[007] At an organizational level there are many units and numerous projects being executed. There are set targets at the organization level that are then cascaded down to first unit and then to project level. Performance data is analyzed at each project, unit and organization level to formulate an action plan for future improvement. Currently, the performance analysis process is reactive but a repetitive one and is based on lag measurement. The performance analysis process cannot prevent the impact on performance indicator at project level before the impact. As a result, none of the above mentioned methods teaches about an improvement in organization performance at project level by evaluating lead measures and enabling the proactive decision making.

[008] Therefore, a solution is required to indicate the impact at project level performance in advance which further enables application of proactive measures in order to prevent or reverse the probable impact and build ability to attain the desired performance target.
OBJECTS OF THE INVENTION
[009] The primary object of the invention is to provide a system and method for predictive analysis towards performance of one or more target objects associated with an organization with respect to one or more parameters affecting the performance.
[0010] The other object of the invention is to determine a proportionality relation by using a logical regression technique, between the performance of the target object and the input parameter.
[0011] The other object of the invention is to provide a conversion of input parameters into one or more group level model factors.
[0012] Yet another object of the invention is to determine an impact of parameters to further depict a probabilistic effect over the proportionality relation to further provide predictive analysis.
[0013] Yet another object of the invention is to determine a delta positive increase or delta negative slippage based on previously obtained customer satisfaction index.
[0014] Yet another object of the invention is to provide a band wise distribution of the impact of parameter to further give a clear visualization of the probabilistic effect.
[0015] Yet another object of the invention is to provide recommendations including best practices and actions with respect to delta negative slippage in a previously obtained customer satisfaction index.
[0016] Yet another object of the invention is to evaluate ability of planned or implemented actions to elevate customer satisfaction index meeting desired target with respect to delta positive increase in a previously obtained customer satisfaction index.

SUMMARY OF THE INVENTION
[0017] The present invention describes a computer implemented system to provide predictive analysis towards performance of one or more target object associated with an organization with respect to one or more parameters affecting the performance. The system comprises a user interface configured to receive input parameters along with an intensity level, and a processing engine coupled to a memory. The processing engine is configured to determine a proportionality relation by using a logical regression technique. between the performance of the target object and the input parameter so selected. The processing engine further comprises a conversion module configured to select a threshold value from a pre-defined truth table to convert the input parameters selected by the user into group level model factors with the associated intensity level and an evaluation module configured to determine an impact of the parameters and further split the impact in order to obtain a band wise distribution. The band wise distribution comprises one or more numerical ranges depicting a probabilistic effect of parameters towards the performance of one or more target object. An output generation module is configured to generate a probabilistic effect of the impact over the proportionality relation so determined to further provide predictive analysis. The output generation module is further configured to generate a standard form of the band wise distribution so obtained and to provide one or more recommending action with respect to the probabilistic effect so depicted.
[0018] The present invention also provides a method performed on a computer to provide predictive analysis towards performance of one or more target object associated with an organization with respect to one or more parameters affecting the performance. The method comprises steps of allowing a user to select input parameters along with an intensity level. The input parameters and the intensity level so selected by the user are processed to determine a proportionality relation by using a logical regression technique. between the performance of the target object and the input parameter so selected. The processing comprises of steps of selecting a threshold value from a pre-defined truth table to convert the input parameters selected by the user into a group level model factors with

the associated intensity level and determining an impact of the parameters and further split the impact of the parameters to obtain a band wise distribution, wherein the band wise distribution comprises of one or more numerical ranges depicting a probabilistic effect of the impact towards the performance of one or more target object. The method further comprises of generating a probabilistic effect of impact of parameter over the proportionality relation so determined to further provide predictive analysis. The method further comprises of generating a standard form of the band wise distribution so obtained and to provide a recommending action with respect to the probabilistic effect so depicted.
BRIEF DESCRIPTION OF DRAWINGS
[0019] Figure 1 illustrates system architecture to provide a predictive analysis in accordance with an embodiment of the invention.
[0020] Figure 2 illustrates a flow chart towards the performance of system to provide predictive analysis in accordance with an alternate embodiment of the invention.
[0021] Figure 3 illustrates a functioning of the system in accordance with an exemplary embodiment of the invention.
[0022] Figure 4 illustrates an overall functioning of invention in accordance with an exemplary embodiment of the invention.
DETAILED DESCRIPTION
[0023] Some embodiments of this invention, illustrating its features, will now be discussed:
[0024] The words "comprising", "having", "containing", and "including", and other forms thereof, are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items.
[0025] It must also be noted that as used herein and in the appended claims, the singular forms "a", "an", and "the" include plural references unless the context clearly dictates otherwise. Although any systems, methods, apparatuses, and devices similar or

equivalent to those described herein can be used in the practice or testing of embodiments of the present invention, the preferred, systems and parts are now described. In the following description for the purpose of explanation and understanding reference has been made to numerous embodiments for which the intent is not to limit the scope of the invention.
[0026] One or more components of the invention are described as module for the understanding of the specification. For example, a module may include self-contained component in a hardware circuit comprising of logical gate, semiconductor device, integrated circuits or any other discrete component. The module may also be a part of any software program executed by any hardware entity for example processor. The implementation of module as a software program may include a set of logical instructions to be executed by the processor or any other hardware entity. Further a module may be incorporated with the set of instructions or a program by means of an interface.
[0027] The disclosed embodiments are merely exemplary of the invention, which may be embodied in various forms.
[0028] The present invention relates to a computer implemented system and method to provide predictive analysis towards performance of target objects associated with an organization with respect to one or more parameters affecting the performance. More particularly, the system and method determines a proportionality relation between the performance of the target object and the parameters affecting the performance to find out the impact of the parameters over the performance. The system and method generate a probabilistic effect of the impact over the proportionality relation so determined to further provide predictive analysis. Further, the system and method provides a standard form of band wise distribution of impact of parameters used to generate a probabilistic effect of the impact over the performance of target object.
[0029] In accordance with an embodiment, referring to figure 1, the system (100) comprises a user interface (102) configured to receive parameters to be fed as an input along with an intensity level. The system (100) further comprises of a processing engine (104) coupled

to a memory (114), the processing engine (104) is configured to determine a proportionality relation by using a logical regression technique, between the performance of the target object and the parameter so selected via user interface. The processing engine (104) further comprises a conversion module (106) configured to select a threshold value from a pre-defined truth table to convert the parameters selected by the user into a group level model factors with the associated intensity level. The processing engine (104) further comprises an evaluation module (108) configured to determine an impact of the parameter fed as the input and further split the impact in a band wise distribution. The system (100) further comprises an output generation module (110) configured to generate a probabilistic effect of the impact of parameters over the proportionality relation so determined. The probabilistic effect provides predictive analysis towards performance of the target objects associated with the organization.
[0030] The user interface (102) is configured to allow the user to select the parameters to be fed as the input along with an intensity level (as shown in step 202 of figure 2). The parameters further comprises of a projects related data, customer satisfaction index information and corresponding customer survey response information. The project related data further comprises of project events, and project scenarios, project issues or weaknesses, along with severity (intensity level). The parameters providing a negative effect comprises of project related issues, events or weakness and the input parameters providing a positive effect comprises of project related best practices, actions or improvements. The user interface (102) receives the parameter that comprises previous probabilistic effect so generated for performing a next predictive analysis. The previous probabilistic effect generated comprises previously calculated Customer Satisfaction Index data for the projects selected so far.
[0031] In accordance with another embodiment, the parameters fed as the input comprises organization customer satisfaction survey data extracted for a defined time period. For example, the data is extracted on a half yearly basis. The projects data is collected for which customer satisfaction index data has been collected. The project related data further includes project management, governance, competence, cost, quality, schedule,

and responsiveness, and resource management, problem solving attitude, value addition and political aspect. Identification of factors related to project weakness or issues etc) could be triggered from various sources that includes health check, RAG (Red Amber & Green) assessment, management review, focus or risk review, audits, customer escalations, verbal dissatisfaction and missed commitments. Population of the project data is further segmented into three sub populations as projects for which satisfaction index is dropped or decreased, projects for which satisfaction index is increased or improved and projects for which satisfaction index remained same. For example, list of 42 factors was finalized from organization experience that primarily influence (negatively or positively) client perception that is reflected in customer satisfaction index. The factors represent various scenarios, events, situations at project level during typical software development and service life cycle.
[0032] According to another embodiment, a survey enabled in a knowledge management system, can be used to capture relevant factors and level of severity from the project data. Survey data captured along with qualitative feedback to have input factors that results in respective projects client satisfaction index slippage or increase.
[0033] The input parameters are selected along with an intensity level wherein the intensity level further comprises intensity of the impact that can influence the target object. For example, the intensity level considered here is low, medium, high and very high. The intensity level is further dependent or is set according various combinations built from the input parameters to further form group models.
[0034] In accordance with another embodiment, the input parameters are pre-processed before the user's selection through the user interface with respect to their affect on the performance. The input parameters are pre-processed by way of operations, the operations further comprises of data formatting, data cleansing, rationalization, transformation, factor grouping and model fitment. The pre-processing may be a single occurring instance before performing predictive analysis of the performance of target objects. The pre-processing helps in filtering or sorting the total parameters to further

retain pertinent parameters which are later selected and fed as the input to the system (100). The pre-processing includes executing a 1st pass to identify groups of the input parameters or factors that have significant impact on the performance of target object. To execute 2nd pass of logistic regression to derive model variables (Constants, Coefficients etc) and formulate logit equation which is further used to calculate the impact of parameters or factors on the performance of target object depicting probabilistic effect in terms of band wise distribution.
[0035] In accordance with another embodiment, the system (100) further comprises the processing engine (104) coupled to the memory (114), the processing engine (104) is configured to determine proportionality relation by using a logical regression technique, between the performance of the target object and the input parameter so selected. The performance of target object further measured in terms of customer or client satisfaction index. The target object further comprises prediction in variation of customer satisfaction index.
[0036] The processing engine (104) further comprises of the conversion module (106) configured to select a threshold value from a pre-defined truth table to convert the input parameters selected by the user into a group level model factors with the associated intensity level (as shown in step 204 of figure 2), The conversion of the input parameters is to provide flexibility to the user and to cater to variation in user selection of factors within a logical group. For example, intensity factor of both individual factors and group variables could be at 4 levels (low, medium, high and very high). An inbuilt rule engine (truth table mapping) converts the user selected input factors along with associated intensity into respective group variables' magnitude or severity. The built-in truth table elevates group level severity (factor ordinal levels) used as input to the evaluation module (108). The processing engine ()104 by way of further modules performs all the calculations by using various techniques/set of embedded instructions.
[0037] In accordance with another embodiment, the factors are grouped based on logical relationship and mutual exclusivity. The 42 factors are grouped and transformed into 11

logical groups. The pre-processing includes preliminary data filtering (of survey feedback) and converting initial 42 factors into set of logical group variables based on mutual exclusivity, coherence and logical relationships. For example, Domain Competence, Technical Competence and Project Management Competence are three distinct factors. Survey feedback indicates that either one (mutually exclusive) of these three factors are selected which are logically grouped (as competence) into single Group Variable. Similarly, the 42 factors were transformed into logically bound 11 group variables (factors), by assigning a suitably elevated intensity, to provide multiple selections within a group variable. Effect of the various group variables to influence the outcome was determined statistically by applying logistic regression. Applying logistic regression technique (1st pass) to key input group variables (7 out of 11) that have critical cause and effect relationship in order to influence the outcome is also identified
[0038] In accordance with another embodiment, the factors fed as input (giving a negative effect) that are converted into groups as presented below:

Factors: Scenario/Event/Issues/Weakness Group
Resource Competence/ Exp level gap—>Domain Grp2
Resource Competence/ Exp level gap—>Project/ Program Management Grp2
Resource Competence/ Exp level gap—>Technical Grp2
Configuration Mgmt →Process not followed Grp3
IT Governance→Code Quality/ Stds compliance Grp3
Non Functional Req→Access Control/ Security Issues Grp3
Non Functional Req→ Performance Issues Grp3
Customer Connect →Lack of Leadership connect Grp4
Project Mgmt - Governance--->Absence of good Metric reporting/ Grp5
dashboard
Project Mgmt - Governance—>Inability to flag risk/ issue well in Grp5
advance
Project Mgmt - Governance—>Lack of Customer connect, Transparent Grp5
sharing, Status review etc
Go-live Performance → Backout /$ Impact /Down Time/ High Cust FTE Grp6
Quality of Deliverables →Go Live-Show-stopper/ High Sev Defects Grp6
Political/ Other→Manager is pro-competitor etc Grp7
Political/ Other→Organization Change in Customer Organization Grp7
Escalation/ Complaint Mgmt—>Issues with Responsiveness/ RCA/ Grp8
Formal & On Time closure

Escalation/ Complaint Mgmt—>Urgency/ priority not shown to customer Grp8
concern/ feedback Grp8
Customer Priority—>Support documentation, User Training etc not
prioritized/ addressed Grp8
Collaboration—>Issues with other entities, vendors/ 3rd parties impacted
Customer/ User GrpS
Preventive VS Reactive—>Lack of focus in CTB (Preventive/ Adaptive
maintenance, Enhancements etc) Grp8
Proactive VS Reactive—>Reactive process/ management Grp8
RCA and Problem Solving—>Root Causal/ Problem solving focus
missing
SIT/UAT - Test Mgmt—>Test Failure/ High Defect Rate/ Show-stopper/ Grp9
High Sev Defects
SIT/UAT - Test Mgmt—>Defects/ Functional Gaps leading to CRs Grp9
SIT/UAT - Test Mgmt—>High Business/ User FTE Grp9
SIT/UAT - Test Mgmt—>Poor Test/ Path/Scenario/ Data Coverage Grp9
Quality of Service/ Deliverables—>Service or Delivery Quality issues Grp9
Resource Mgmt→Attrition of Key/ named resources GrplO
Resource Mgmt→Resource Availability/ On-boarding issue, inability to GrplO
ramp up GrplO
Resource Mgmt → Resource Turnover/ Release w/o customer consent GrplO
Resource Mgmt → Shifting key resources from Onsite
Schedule adherence →Delay in Intermediate Deliverables Grpil
Schedule adherence →Shifi7 postponement in Release/ Go-live Milestone Grpll
Value Addition →Contractual savings/ value add not met Grpl2
Value Addition → No value add apart from BAU Grpl2
Value Addition → Proactive ideas/ Out of Box thinking not shared Grpl2
Table 1
[0039] In accordance with an exemplary embodiment, the input factors (giving a positive effect) converted into groups is presented below:

Factors: Action/Improvement/ Strength/ Appreciation Group
Competence (Proj Mgmt)—> Knowledge/ Skill Grpl
Competence (Technology)—> Knowledge/ Skill Grpl
Competence (Domain)—> Knowledge/ Skill Grpl

Project Mgmt/ Governance →Proactive sharing of issues, flagging Risks
Grp2
Project Mgmt/ Governance→Detailed planning & regular sharing of progress & Grp2
status
Project Mgmt/ Governance →Regular connect at project, account & leadership Grp2
level
Project Mgmt/ Governance→Regular review of Project performance by account Grp2
leadership
Escalation/ Complaint Mgmt→No complaint but appreciations, received Grp3
Customer priority→Implicit customer priority (documentation, training etc) has Grp3
been prioritized
Proactive Vs Reactive→Proactive actions, planning, thought process played as Grp3
key differentiator
Preventive Vs Reactive→Focus on preventive support helped reducing need for Grp3
reactive support
RCA and Problem Solving→Root cause and problem solving focus made Grp3
substantial difference
Collaboration →With other entities to prioritize meeting project objectives/ Grp3
performance baselines
Political/ Other→ Manager is pro-competitor etc Grp4
Political/ Other→ Organization change in customer organization Grp4
Go-live/ Release Performance→Ability to contain High Severity defects/ show Grp5
stoppers/ Business impact
SIT/UAT- Test Mgmt→ High pass rate, no major defects/ show stoppers& Grp5
backlog
SIT/UAT - Test Mgmt—> Req/ Design Gaps not found during SIT/UAT Grp5
Quality of Deliverables→ High quality deliverables maintained all through Grp5
Quality of Deliverables →Containment of high Severity defects/ show shoppers Grp5

in SIT/UAT

KPI (Metric/SLA) Performance→ Well within customer expectation, showing Grp5
improvement trend
IT Governance → Compliance to IT framework/ governance Grp6
IT Governance → Meeting code quality expectations Grp6
Non Functional Requirement → Meeting performance and security expectations Grp6
Resource Mgmt→ No customer impact (induction, on-boarding, ramp-up, Grp7
sudden release or attrition)
Value Addition → Sharing ideas/ Suggestions/ Improvements/ thought Grp8
leadership/ best practices etc
Usage of Delighter→Tool Usage, Reusable components, Best practice adoption etc Grp8
Schedule adherence → did not include any delay in overall completion and major Grp9
milestones
Schedule adherence→ critical paths were managed effectively Grp9
SIT/UAT- Test Mgmt→ Schedule compliance Grp9
Configuration Mgmt→ No surprise from configuration lapses Grp10
Change Mgmt→No Customer impact (Budget overrun and/or CRs due to Grp10
recruitment gaps)
Table 2
[0040] In accordance with anotherembodiment, builtin truth table and theusage to determine group level severity from selected factors severity/ magnitude is presented below:

Level 1 - Truth Table Group Severity Action
If None entered ZERO(0) EXIT
Atleast 1 Low(l) Low(l)
Atleast 1 Med(2) Med(2)
Atleast 1 High(3) High(3)
Atleast 1 Very High(4) Very High(4) EXIT
Level 2 - Truth Table Group Severity
if All Low(l) Low(l) EXIT
If <=2 Low(l) & All other Not Entered(O) Low(l) EXIT
If >2 Low(l) & All other Not Entered(O) Med(2) EXIT
If 1 Med(l) & All other Low(l) or Not Entered(O) Med(2) EXIT
If 1 High(l) & All other Low(l) or Not Entered(O) High(3) EXIT
If 1 or 2 Med(2) & All other Low(l) or not entered(O) Med(2) EXIT
If 3 Med(2) & All other Low or Not entered High(3) EXIT
If>3Med(2) Very High(4) EXIT
If >=2 High(3) Very High(4) EXIT
If 1 High(3) and >=2 Med(2) Very High(4) EXIT
If l High(3)and>=3Low(l) Very High(4) EXIT
If 1 High(3) and 1 Med(2) and 2 Low(l) Very High(4) EXIT

Table 3

[0041] Referring to figure 1, the system (100) further comprises of the evaluation module (108) configured to determine an impact of parameters in terms of the band wise distribution (as shown in step 206 and 208 of figure 2). Further the band wise distribution comprises one or more numerical ranges depicting probabilistic effect of the impact of parameters towards the performance of one or more target object. The ranges of the band wise distribution of probability of the of parameter comprises of 0-2%, 2-5%, 5-10%, 10-15%, 15-20% and %20+.
[0042] Referring to Figure 3, a target is set for Client Satisfaction Index at organization level based on history information, experiences and leadership mandate, often termed as expected Process Performance Baseline (PPB). The underlined process is Organization Performance Management Process which is linked to other sub-processes. While setting

up the target, the evaluation of current process performance is performed to assess current capability. The evaluation is further linked to various units' performance objectives. captured at individual client touch-point (executing projects).In general. Client Satisfaction is captured by conducting surveys, measuring the survey, analyzing and formulating action plan aimed at future improvement.
[0043] Referring to figure 3, by way of a specific example, the system (100) may work as a process performance model to perform organization's performance management by capturing key objectives (input parameters / Factors) and determining Client Satisfaction Index. There may be plurality of units and projects associated with them. The processing engine (104) processed the input parameters / Factors to evaluate parameter and to provide predictive analysis. For example, the processing engine (104) evaluates ability to meet Project Level Performance target (PPB (as shown in Figure 3). Process Performance Baseline (PPB) is process performance target in terms of Customer / Client Satisfaction Index.
[0044] According to another embodiment, the satisfaction index captured from survey and the calculated impact on the satisfaction index providing delta satisfaction index is a continuous data. So it has been considered as various bands (0-2%, 2-5%, 5-10%, 10-15%, 15-20% and %20+) to convert the outcome data that is the parameter probability data (probabilistic effect of the impact of parameters over the proportionality relation) as discrete (as shown in step 210 of figure 2).
[0045] The evaluation module (108) is further configured to apply logistic regression technique to establish a cause and effect relationship (by way of proportionality relation) between group level model factors with associated intensity level and its effect on one or more target object. The evaluation module (108) further identifies key influence factors (p value), relative importance / Weightage (odds ratio) of various factors, probability distribution in various bands and model accuracy/ fitment (concordance) by using the group level model factors with associated intensity level obtained from the conversion module (106). Further, logistic regression technique (2nd Pass) is executed to derive

model variables (Constants, Coefficients etc) and formulate Logit equation to further calculate the band wise distribution of impact of parameter and to further generate a probabilistic effect of the of parameters over the proportionality relation so determined in order to provide the predictive analysis.
[0046] The calculative part so performed by the processing engine (104) and later the output generation module (110) is presented below:
Logit(P) = Gj=B0 + Bl*Xl + B2*X2 + + Bk*Xk
P = Probability (instance j) = 1 / (1+Exp [-Gj])
[0047] Still, referring to system (100) and the evaluation module (108), the proportionality relation between the performance of the target object and the input parameters further provides decrease in the performance of one or more target object or an increase in the performance of the target object with respect to the input parameters. The evaluation module (108) determines the proportionality relation to further provide at least a delta positive increase or a delta negative slippage in previously obtained customer satisfaction index as input parameter.
[0048] Herein, the parameters when fed as the input provides a negative effect comprises of project related scenarios, issues, events or weakness that has visibility to client and the input parameters providing a positive effect comprises of project related best practices, actions or improvements visible to client.
[0049] According to yet another embodiment, the system (100) is configured to establish a proportionality relationship to predict possible slippage (delta decrease) in client satisfaction index (CSI) for a set of applicable factors and level of influence. Delta CSI slippage calculated for each project instance. In order to provide better predictability to user based on range of slippage, 6 possible bands chosen. Accordingly, delta CSI slippage data converted into six possible bands (0-2%, 2-5%, 5-10%, 10-15%, 15-20% and >20%). Fitment in a band containing a higher slippage probability, greater is the risk, which in turn warrants management attention and rigor in action planning and

monitoring. Similarly, the system (100) is also configured to establish a relationship to predict possible rise (delta increase) in client satisfaction index (CSI) for a set of applicable factors and level of influence. The objective of predicting delta increase in CSI for a set of applicable factors and level of influence is to evaluate ability of planned/ implemented actions to elevate CSI level in meeting desired target.
[0050] Still referring to figure 1, the system (100) comprises the output generation module (110) configured to generate probabilistic effect of the impact of parameter over the proportionality relation so determined to further provide the predictive analysis (as shown in step 210 in Figure 2). The output generation module (110) is further configured to generate standard form of the band wise distribution so obtained from the evaluation module (108). The output generation module (110) thus configured has builtin ability to rationalize the band wise distribution obtained from evaluation module based on previous customer satisfaction index data. The standard form of the band wise distribution is generated by applying a technique of normalization.
[0051] According to another embodiment, the output generation module (110) has builtin ability to rationalize the band wise probability distribution based on previous customer satisfaction index data. The rationalization is performed to transform the model output into more realistic bands when organization previous experience data is applied. The output generation module (110) normalizes the probability distribution based on current satisfaction index bands (For example as 90-100%, 80-90%, 70-80%, 60-70%, <60% etc). The normalization is performed in order to rationalize the model further based on a project's current satisfaction level. It has been observed this plays an important role in determining the delta negative or positive i.e. client satisfaction decrement or improvements, when similar input factors is chosen. This is calculated by using a formula:
Normalized (independent) Probability (band 1) = Probability (band 1| organization experience) * Probability (band 11 selected model factors)

[0052] Still, in accordance with another embodiment, the output generation module (110) is further configured to provide one or more recommending action with respect to the probabilistic effect so depicted. The output generation module (110) further comprises an alert module (112) configured to provide recommendations with respect to delta negative slippage so determined. The output generation module (110) is configured with an action knowledge base that comprises organization best practices to improve the factors modeled in the system. Best practices or actions based on previous data will be suggested for corresponding delta negative slippage. When the user selects a combination of input factors, corresponding best practices (suggested improvement actions) would be guided.
[0053] Yet, according to another exemplary embodiment, the system (100) uses present scenarios of a project and predicts probability of possible (delta) slippage in various bands. This helps projects to link project events/ scenarios to probable impact (-ve) on future client satisfaction index, by observing the probability distribution in various bands and use it as lead indicators to act proactively and take informed decision towards reversing or minimizing the impact. The model also normalizes the probability distribution based on current client satisfaction index band (90-100%, 80-90%, 70-80%, 60-70%, <60% etc). The model also suggests a set of best practices/ actions, based on selected weakness (model input factors) that was captured through a similar survey among projects, for which there was %increase in client satisfaction index.
[0054] The system and method illustrated provides predictive analysis towards performance of one or more target object associated with an organization with respect to one or more parameters affecting the performance, the system and method may be illustrated by a working example stated in the following paragraph; the process is not restricted to the example only:
[0055] In accordance with another embodiment, the system (100) is used to provide predictive analysis towards increase or decrease in customer satisfaction index (which is the target object here). The system (100) may be configured in manner to work as negative model (-ve) when there is decrease in Client Satisfaction Index and positive

model (+ve) when there is increase in Client Satisfaction Index based on the input parameters. The system (100) is also used in variety of scenarios as explained below.
Case 1: Any project, at any point of time, having specific issue(s) or weakness(s) (as
input parameters and relevant intensity level selected by the user)
i) Use -ve Model to assess possible impact on CSI (delta %decrease in future CS1)
Case 2: Any project, having specific issue(s) or weakness(s) and some strength or improvements (as input parameters and relevant intensity level selected by the user)
ii) Use both -ve Model and +ve Model to assess possible impact on CSI, but addition of probabilities is not recommended
Case 3: Projects fail to attain desired CSI (Target/ Specification Limit) or if CSI is dropped or having dissatisfaction (Attribute or as mentioned in Top 3 OFI - Opportunity for Improvement section); must have action plan ready and available
iii) Use -ve Model
a) To validate, if planned action(s) are in line with suggested action(s)/ Best Practices
iv) Use +ve Model
a) To validate how much %CSI elevation would be possible by implementing these action(s) and if that is sufficient to meet the Target
b) Ongoing basis, to evaluate effectiveness of these implemented actions and probable +ve influence on CSI
Case 4: Projects with High CSI and only Strengths (no issue, no weakness)
v) To raise the bar and use +ve Model based on further improvement areas selected and acted upon
[0056] Sample illustration of CSS Negative Model to demonstrate how input factor selection is translated into model output as probability distribution in 6 bands:

Factor selection by User
Intensity Value Group Elevated Group Variable after truth table applied
Resource Competence/Exp level gap High 3 Grp2 High 3
→ Project / Program Management
Project Mgmt - Governance → Medium 2 Grp5 Medium 2
Inability to flag risk / issue well in
advance
SIT/UAT - Test Mgmt—>Test High 3 Grp9 Very High 4
Failure/ High Defect Rate/ Show-
stopper/High Severity Defects
SIT/UAT - Test Mgmt → High Medium 2

Business / User FTE
SIT/UAT - Test Mgmt →Poor Test Medium 2

/Path / Scenario /Data Coverage
Note: No factors selected in Other Groups Grp7, Grp8, Grpl 0 and Grp 2.
Table 4

User Selection of Model Factors
Grp2 Grp5 Grp7 Grp8 Grp9 GrplO Grpl2
3 2 0 0 4 0 0
Table 5

Logistic Regression Model Equations for CSS -ve Model
Coeff(Grp2) Coeff(Grp5) Coeff(Grp7) Coeff(Grp8) Coeff(Grp9) Coeff(Grpl0) Coeff^Grpl2)
-0.4910 -0.3464 -0.5766 -0.3183 -0.5092 -0.4093 -0.2192
Const 1 Const2 Const3 Const4 Const5
-0.8071 0.7106 2.1771 3.0233 4.0331
Table 6
Logit(P) = Gl = G(Bandl) = Const 1 + [Grp2 * Coeff(Grp2) + Grp5 * Coeff(Grp5) + Grp7 * Coeff(Grp7) + Grp8 .* Coeff(Grp8) + Grp9 * Coeff(Grp9) + GrplO * Coeff(GrplO) + Grpl2 *Coeff(Grp 12)] =-5.0097

P = Probability (Bandl) = 1 /(1+Exp [-G(Bandl)]) = 0.006628725
P1 = Probability (Band 1) = 0.006628725
Logit(P) = Gl = G(Band2) = Const2 + [Grp2 * Coeff(Grp2) + Grp5 * Coeff(Grp5) + Grp7 * Coeff(Grp7) + Grp8 * Coeff(Grp8) + Grp9 * Coeff(Grp9) + GrplO * Coeff(Grp 10) + Grpl2 *Coeff(Grpl2)] =-3.4920
P = Probability (Bandl&2) = 1 / (1+Exp [-G(Band2)]) = 0.02954] 173
P2 = Probability (Band2) = 0.029541173 - 0.006628725 = 0.022912448
Logit(P) = Gl = G(Band3) = Const3 + [Grp2 * Coeff(Grp2) + Grp5 * Coeff(Grp5) + Grp7 * Coeff(Grp7) + Grp8 * Coeff(Grp8) + Grp9 * Coeff(Grp9) + GrplO * Coeff(Grpl0) + Grpl2 *Coeff(Grpl2)] =-2.0256
P = Probability (Bandl,2&3) = 1 / (1+Exp [-G(Band3)]) = 0.116546327
P3 = Probability (Band3) = 0.116546327 - 0.029541173 = 0.087005155
Logit(P) = Gl = G(Band4) = Const4 + [Grp2 * Coeff(Grp2) + Grp5 * Coeff(Grp5) + Grp7 * CoerT(Grp7) + Grp8 * Coeff(Grp8) + Grp9 * Coeff (Grp9) + GrplO * Coeff(GrplO) + Grpl2 *Coeff(Grp 12)] =-1.1793
P = Probability (Bandl, 2, 3 &4) = 1 / (1+Exp [-G(Band4)]) = 0.235174484
P4 = Probability (Band4) = 0.235174484 - 0.116546327 = 0.118628157
Logit(P) = Gl = G(Band5) = Const5 + [Grp2 * Coeff(Grp2) + Grp5 * Coeff(Grp5) + Grp7 * Coeff(Grp7) + Grp8 * Coeff(Grp8) + Grp9 * Coeff(Grp9) + GrplO * Coeff(GrplO) + Grpl2 *Coeff(Grp12)] =-0.1695
P = Probability (Band 1,2,3;4&5) = 1/(1 +Exp [-G(Band5)]) = 0.457726163
P5 = Probability (Band5) = 0.457726163 - 0.235174484 = 0.222551679
P6 - Probability (Band6) = 1 - (P1+P2+P3+P4+P5) = 1 - 0.457726163 = 0.542273837

[0057] Below are listed the output results (i.e. standardized probabilistic effect in terms of numerical values for both 2the models:
Case 1: Band wise probability distribution when last received customer satisfaction level (CSI) is 80-90%

Population Band Dist. %
CSl-% slippage Band Core model Probability CSK80-90%> Conditional Probability Normalized Model Output
1 0-2% 0.006628725 0.251552795 0.001667474 0.015845874 1.58%
2 2-5% 0.022912448 0.251552795 0.00576369 0.054771883 5.48%
3 5-10% 0.087005155 0.223602484 0.019454569 0.184875195 18.49%
4 10-
15% 0.118628157 0.118012422 0.013999596 0.133037031 13.30%
5 15-
20% 0.222551679 0.62111801 0.013823086 0.131359666 13.14%
6 20+% 0.542273837 0.093167702 0.050522407 0.480110352 48.01%
Table 7
Case 2: Band wise probability distribution when last received customer satisfaction level (CSI) is 60-70%

Population Band Dist. %
CSI-% slippage Band Core model Probability CSK60-70%> Conditional Probability Normalized Model Output
1 0-2% 0.006628725 0.105263158 0.000697761 0.006898963 0.69%
2 2-5% 0.022912448 0.210526316 0.004823673 0.04769307 4.77%
3 5-10% 0.087005155 0.421652632 0.036633749 0.362208608 36.22%
4 10-15% 0.118628157 0.157894737 0.018730762 0.185196525 18.52%
5 15-20% 0.222551679 0.052631579 0.011713246 0.115812296 11.58%
6 20+% 0.542273837 0.052631579 0.028540728 0.282190539 28.22%
Table 8
[0058] As shown in above table 7 and 8, the effect of output generation module (110) is presented. In accordance with another exemplary embodiment, the system (100) is a core model. The system (100) or Core model calculated probability is further normalized (conditional probability) with overall distribution probability to predict band wise probability distribution for selected factors. The system (100) outputs maximum probability in band 6 (54.22%) and it cannot further optimize this based on previously obtained CSJ Band. Hence, for both CSI bands (80-90% and 60-70%) the core model output remains the same. But when effect of output generation module (110) applied the most probable band changes; for CSI band 80-90%, it is Band 6 with probability 48.01 % (normalized), while for CSI band 60-70% it now would show Band 3 with probability 36.22%o (normalized).

WE CLAIM:
1. A system to provide a predictive analysis towards performance of one or more target objects associated with an organization with respect to one or more parameters affecting the performance, the system comprising:
a user interface configured to receive the parameters and an associated intensity level as an input to predict the performance of the target object;
a processing engine coupled to a memory, the processing engine configured to determine a proportionality relation between the performance of the target object and the parameter so received by applying a logical regression technique over the input, the processing engine further comprising:
a conversion module configured to select a threshold value from a pre-defined truth table to convert the parameters so received into a group level model factors with the associated intensity level;
an evaluation module configured to determine an impact of parameters over the performance in terms of a band wise distribution; and
an output generation module configured to generate a probabilistic effect of the impact over the proportionality relation so determined to further provide the predictive analysis.
2. The system of claim 1, wherein the target object further comprises prediction in variation of customer satisfaction index.
3. The system of claim 1. wherein the parameters received by the user interface further comprises projects related data, prior customer satisfaction index information and corresponding survey response information, project events and project scenario.
4. The system of claim 1, wherein the input received by the user interface further comprises a previously generated probabilistic effect to be used for performing a next consecutive predictive analysis.

5. The system of,claim 1, wherein the evaluation module splits the variance of the parameters in the band wise distribution, the ba::d wise distribution comprises numerical ranges of 0-2%, 2-5%, 5-10%, 10-15%, 15-20% and %20+
6. The system of claim 1, wherein the processing engine determines the proportionality relation to further provide at least one of a delta positive increase, or a delta negative slippage in a previously obtained customer satisfaction index.
7. The system of claim 1, wherein the output generation module further comprises an alert module configured to provide recommendations with respect to delta negative slippage so determined.
8. The system of claim 1, wherein the input parameters received are further categorized as parameters inducing a negative effect and parameters causing positive effect.
9. The system of claim 1, wherein the output generation module is further configured to generate standard form of the probabilistic effect by applying a technique of normalization over the band wise distribution; the standard form is further generated by using a previous customer satisfaction index data.
10. A computer implemented method for predictive analysis towards monitoring performance of one or more target object associated with an organization with respect to one or more parameters affecting the performance, the method comprising steps of: receiving the parameters along with an intensity level as an input to predict the performance of the target object;
processing the input parameters to determine a proportionality relation between the performance of the target object and the input parameter so received by applying a logical regression technique over the input, the processing comprising steps of:
selecting a threshold value from a pre-defined truth table to convert the parameters so received into a group level model factors with the associated intensity level;
determining an impact of parameters over the performance, wherein impact is further categorized to obtain a band wise distribution; and generating a probabilistic effect of the impact of parameter over the proportionality relation so determined to further provide the predictive analysis.

11. The method of claim 10, wherein the target object further comprises customer satisfaction index prediction. .
12. The method of claim 10, wherein the parameters so received further comprises projects related data, customer satisfaction index information and corresponding survey response information, project events and project scenario.
] 3. The method of claim 10, wherein the input so received further comprises a previously generated probabilistic effect to be used for performing a subsequent predictive analysis.
14. The method of claim 10, wherein a numerical range of the band wise distribution of the impact comprises of 0-2%, 2-5%, 5-10%, 10-15%, 15-20% and %20+.
15. The method of claim 10, wherein the proportionality relation is further indicative of at least one of a delta positive increase, or a delta negative slippage in a previously obtained customer satisfaction index.
16. The method of claim 10, wherein the parameters providing a negative effect comprises of project related issues, events or weakness and the parameters providing a positive effect comprises of project related best practices, actions or improvements.
17. The method of claim 10, wherein the probabilistic effect obtained from band wise distribution is generated in a standard form by applying a technique of normalization over the band wise distribution, the standard form is further generated by using a previous customer satisfaction index data.
18. The method of claim 10, wherein the output so generated further provides recommendations with respect to the delta slippage of customer satisfaction index so determined.