4.

Scope of Work and General Requirements

4.1. Scope of work
4.1.1. Introduction
Assam Power Distribution Company Limited (APDCL) is a public limited
company wholly owned by the Government of Assam. It was incorporated
on the 23rd day of October’2009 and has been registered under Indian
Companies Act’1956.
The primary purpose of the Company is to undertake distribution, trading
and supply of electricity in the state of Assam in accordance with
provisions of applicable law and all activities ancillary or appurtenant
thereto. It has also the mandate to develop, maintain and operate the power
distribution system in the state of Assam. In carrying out the work of
supplying power, APDCL reaches every part of the state. The company is
serving the people of Assam with a consumer base of 64 lakhs.

4.1.2. Consolidated Scope of Work

a. Preparation of detailed implementation plan covering the complete life
cycle of the project.
b. Complete project management activities as per standard norms & practices.
c. Provisioning of cloud Infrastructure & associated Cloud software and
services(VM’s, connectivity, storage, security, etc) of suitable configuration
required for all working environments during the contract period.
d. Deploying applications on Cloud, user administration, security
administration, planning and implementation of Cloud Management and
Monitoring Portal for complete infrastructure and services procured.
e. Supply, Install, Configure, Customize and implement offered Data
Management, Analytics and BI Solution using COTS based software as per
requirements.
f. Bidder needs to ensure Quality assurance, compliance and security as per
standards.
g. The Solution shall support future requirements of APDCL including GIS
(Geographical Information System) & Renewable Analytics.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 24 of 115
 h. Post-go-live support for the system for product, functional & technical
support for implemented solution & related services.

4.1.3. Below is the consolidated functional SOW

 Reduce Revenue lost due to Non-Technical Losses
 Accurately forecast Day-Ahead Demand
 Optimization of Peak Load
 Effective Predictive Maintenance
 Launching more Effective Demand Response Programs
 Enable or Improve efficiency of desk investigations (time per alert/lead
investigation)
 Recoup investment in IT/OT systems and smart meters through increased
identification and prevention of fraud
 Identify customers who are switching/dropping to avoid paying other
delinquent accounts
 Provide AI Enabled Decision Support System to Department
 To provide BI and Reporting tools for generation of MIS, Reports,
Dashboards to help DISCOM Officers and Management in taking decisions
for reducing losses, optimizing operations and enhancing customer
satisfaction etc. All relevant officials shall use the system. Custom query
report generation functionalities as per user requirements of configuring
different KPIs in the system should also be available without constraint on
the number of custom reports through the system. It is the aim of APDCL to
be able to derive actionable insights using analytics.

4.2. Detailed Scope of Work

4.2.1. Data Integration

The solution should provide the capability to natively connect to the
various disparate sources as mentioned below in sub section A and
underlying operational systems of data. The solution should perform
transformations of data from native sources to the new target system

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 25 of 115
 repository which shall be deployed as part of this project, without the
underlying operational systems needing to be changed.

As a first step to gather insights from the data, it is important to

integrate the voluminous data from various databases or sources into a
single format to get a holistic view of the functioning of entire process.
With the help of the data integration solution capabilities, APDCL
intends to integrate data from various sources as mentioned below in
sub section A, perform various transformations, aggregations,
validations and ability to handle voluminous amount of data and
process data in near-real-time and in batch mode using ETL
functionalities. Also, utilize ETL/ELT functionalities to process the
various jobs to automatically update the data within data warehouse
and refresh reports and analysis once new period data is available into
the source database.
A comprehensive data integration component shall help to:

a. Extract/Integrate data from all relevant sources

b. Create a complete data model of all entities and their respective
attributes, which will then flow into a data repository for further
analytics and visualization
c. Perform Data Quality Analysis and Imputation
d. Validate and cleanse, de-duplicate data based on Customizable rules

A. Data Sources to be integrated with following Analytics Solution:

1. SCADA system
2. AMI Data with 15 min granularity
3. APDCL Revenue Management System (ARMS)
4. ERP System
5. Data from Multiple Payment Channels
6. Customer Statements
7. SLDC

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 26 of 115
4.3. Data Quality Analysis
As data would be integrated from multiple database sources which would
have different schema/structure, it is important to cleanse, validate and
standardize the data from various sources. Perform various cleansing,
standardization, data profiling analysis, such as structure discovery, data
discovery, relationship discovery to determine the quality state of data and
identify anomalies in data. Solution also facilitates identifying redundancies
and duplicate entry within Master and Transaction data.

4.4. Revenue Assurance and Reconciliation

To meet the core Revenue Assurance and Reconciliation requirements, the
following approach is to be adopted:
 Identify key elements of the billing, payments and Customer
Systems and the main data elements/ variables which are relevant
to the above problem
 Consolidate the Data together and create a master record which
integrates all aspects and measures from these systems.
 Once the data is integrated via an ETL platform, Reconciliation
Platform shall carry out execution on the data to identify, whether the
amount paid has been
o credited
o credited in full in the APDCL customer accounts
Cases where there’s a breach should be alerted by the system for
Investigation and Triage.
 For cases where reconciliation fails, the platform will identify the
possible stage in the process where leakage/ discrepancy happened. A
corrective action by the Department will then be taken to eliminate
Revenue leakage progressively hence curbing illicit behaviour or
processes.
 Collection Reconciliation: The Analytics framework for non-technical
loss addresses the following:

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 27 of 115
 o Enhance information credibility by integrating disparate data
sources from payment, billing, meter data regardless of format
(including unstructured text, like notes fields in service call centers
and bill collection centers) – and applying embedded data quality
techniques to ensure accuracy.
o Find irregularities, deploy rules / control points and build hypothesis
to detect meter errors and understand revenue leakage patterns.
Create benchmark analysis of neighborhood meters, same category
meters.
o To understand usage behavior, it has time series data mining
capabilities that would provide time series decomposition, feature
extraction and dimension reduction capabilities to provide
meaningful customer segments associated with non-technical losses.
o Understand payment behavior and build predictive scores for
customer associated with delayed payment, customer in vulnerable
section, customer with chances of default. These scores can be
utilized for altering communication such as reminders, follow-ups
for collection optimization.
o Improve investigator efficiency that lets investigators capture the
field inspection details against confirmed / suspect theft meters
including photographs, document reports etc. Visualize data from
multiple angles, perform interactive queries and generate inspection
efficiency reports.
o Measure program performance as you define and monitor key
performance metrics via a dashboard environment to determine how
effective and efficient your programs are at avoiding fraud losses.
 Bidder should Identify key elements of the billing, payments and Customer systems
and the main data elements/ variables which are relevant for revenue reconciliation
 System should collect data from all existing systems like Billing Applications,
Customers, Payment Gateways, web applications, wallets and all future systems.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 28 of 115
 System should collect all payment events data including cash, cheque, payment
gateway, wallets etc.
 Solution shall be able to extract data from sources (Core Systems, Excel Files or any
other file formats) transform them, monitor and enhance their quality.
 Consolidate the Data together and create a master record which integrates all aspects
and measures from these systems. Complication lies in granularity of the data
available from these systems and the system should be able to consume and churn
data at each granularity levels.
 Once the data is collected, system shall process and analyze the same to identify,
frauds, risks, whether the amount paid has been credited in full in the APDCL
customer accounts.
 Where reconciliation fails, system shall identify the possible stage in the process
where leakage/ discrepancy happened.
 Any alerts that get generated as a result of these analytical techniques shall be handled
through an Alert Management system.
 Prioritization and queuing of alerts shall be available, and the investigator shall get to
analyze individual alerts as per supporting evidence.
 The investigator should be able to easily qualify the alert and determine whether it
needs to become a suspected case.
 APDCL Staff should be able to carry out necessary action to proceed against or
further monitor the suspect concerned.
 System shall allow management and resolution of different suspected case types.
 System shall support role-based access and interfaces for all types of users and
centralized administration.
 Each user should be provided with customized reports and dashboards.
 To develop various mobile applications which would need to replicate system
functionalities for on the go processing like Bill payment App etc.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 29 of 115
  Detailed General, Technical and Functional Requirements are at Section 10-
“Technical Specifications” to this RFP.

4.5. Demand Forecasting

Proposed Demand forecasting should produce a repeatable, scalable, traceable
and defensible energy forecast performance for planning horizons that range
from day ahead, short-term to long-term. APDCL is expecting the following
functionalities to be available as part of this project:
 Short term, medium term and long term demand forecasting
covering horizon from 5 minutes, 15 minutes, hourly, day ahead
up to 5 year period.
 Granular level forecasting at Network, Organization and
Geographical level hierarchy
 It should be possible to generate macro level forecasts from the
granular level forecasting.
 The demand forecasting system must consider configurable
variables like historical data, category wise load data, season, weather,
special day, calendar, population growth, GDP growth etc. for
preparing the forecast model catering to each granular and macro level.
 User should be able to stress/define scenarios across multiple
parameters.
 Forecast error measurement and feedback mechanism to update
the model based on feedback must be part of the forecasting
system.
 Granular level of forecasting must be performed as a hierarchy

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 30 of 115
 from system level, feeder level, transformer level and capability up to
meter level. This will be useful for selection of the pricing strategy,
distribution-asset planning, capacity augmentation etc.
 Create the Decomposition of electricity load Demand into lower
granularity of time and geo dimension. In order to facilitate decision
making for surplus/shortage, the long term forecast then further needs
to be decomposed at lower level granularity.
 Annual into seasonal: Incorporates the shift effects across season
o Seasonal into monthly: Incorporates the shift effect across
month for any given season
o Monthly to weekly: Incorporates the shift effect across weeks if
any for any given month
o Weekly to daily: Incorporate shift effect across the days for any
given week: for example, weekdays are different from the
weekly.
o Daily to hourly load shape: Incorporate shift effect across
different hours within a day to incorporate the shifts like peak
and off-peak hours.
 Day Ahead Short-Term Demand Forecasting.
o Feeder Level Load Forecasting for the APDCL across all sub-
stations.
o Configuration of the load forecasting solution for day ahead 15
min interval. It should include configuration of the diagnostic
process for baseline forecasting and adding effects such as
Calendar (Weekend Effect), Holiday Effect, special event /
festival effects / shutdowns etc.
o Configure the error analysis and outlier detection.
o Configure the scenario planning for changes in events,
explanatory variables.

4.6. Scenario Analysis

 System should be able to generate short, medium& long-term scenarios
for both demand & supply
o User should be able to stress/define scenarios across multiple parameters
o Effect of different scenarios is required to be factored for short, medium
term and long term forecast as mentioned below:
o Delayed/Early/Normal Rainfall

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 31 of 115
 o Recession
o Migration of Consumer (Hi/Low)
o Open Access Consumer
o Changes in Policy and Regulations
o Change in Tariff (Sensitivity to Tariff)
o Any other scenario which will help in effective planning.
o Non availability of declared capacity
o Transmission constraints etc.
o Covid Lockdowns
o Natural Calamities like flood, earthquake, storm etc.

4.7. Demand Supply Position Map

 The Analytics Solution must generate Daily Power Position Report
highlighting the current portfolio status and availability of Power.
 The system should be able to consider forecasted demand and available
sources of power to create surplus/deficit positions.
 System should also be able to consider diverse scenarios for power availability
to generate position map.
 Based on the forecast generated by the system, and the current contracts, the
system should be able to generate and evaluate the options to select Quantum
for a period up to 30 days
o 15 min/ Monthly Purchase / Sale Quantum
o Peak/off-peak load Purchase / Sale Quantum
o LDC (Load duration Curve– Surplus Shortfall position)
o Shortfall/Surplus analysis
o PLF(Plant Load Factor) Maintainability of the Quantum if it is
contracted
o Rate at which it has to be contracted
o Penalty/Compensation
o Hourly schedule of power w.r.t Working days / Holidays

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 32 of 115
4.8. Enterprise Visualization
The Analytics solution must have Data visualization capabilities that helps
APDCL to explore and make sense of department data. Adding analytics to
their visualizations helps uncover insights buried into data. Department shall
quickly recognize outliers that may affect the Business Operations of the
Utility. The solution should also provide alerts based on user defined criteria
which would be dispensed in the form of emails, portal alerts etc. The solution
should also provide power users the capability to perform ad-hoc analysis on
the data and distribute the results on mobile tablet devices for on-the-go
analysis offline.

Some of key Requirements which APDCL would like to focus to study based on
Visual Analytics:
 Report stating the load forecast for next day in hourly buckets (or sub-
hourly) should be made available before 10 AM on the day before
 Basis this load forecast, open access off take and generation schedule,
another report stating exact trading strategies for APDCL everyday
should be made available
 The solution should provide power position report showing deficit /
surplus under different positions
 The solution should provide pre-optimized and post-optimized scenarios
for power purchase. In the pre-optimized scenario, the portfolio mix,
missed opportunities should be identified
 The solution should be able to support use cases related to smart meters.
 Dashboards highlighting key KPI’s to be monitored by the department
 Customized Dashboards to cater varying business needs of APDCL.

4.9. Data Security Requirements

 The data and applications in the cloud shall be exclusive property of APDCL
and ensuring the security, confidentiality and integrity of the data hosted on
the cloud is the sole responsibility of the bidder. Without the permission of

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 33 of 115
 APDCL they should not be accessible to any other outsider and the bidder
has to ensure this aspect.
 The bidder shall indemnify the owner against any breach of data security and
confidentiality. Security requirements of proposed solution and its
implementation is part of bidder’s scope.
 The Bidder shall follow APDCL’s information security policy and
instructions in this behalf to the extent possible. The policy document will be
shared with the successful Bidder.
 The successful bidder shall ensure that the security best practices provided by
OEM are implemented in the services provided to APDCL.
 The bidder shall propose WAF and DDoS service from the appointed
BIDDERS to secure the implemented solution.
 The Bidder is required to submit Vulnerability Assessment and Penetration
Test (VAPT) reports bi-yearly from a CERT-In empanelled auditor.
 Before operational acceptance of the system, the bidder has to submit latest
VAPT audit report conducted by a CERT-In empanelled auditor.

4.10. Testing & Sign Off Process:

Before Go-Live, the bidder should follow all required testing methodologies
and its lifecycle. Bidder should submit test case documents before start of the
test.
- Unit Testing
- Integration Testing
- QA Testing
- User Acceptance Testing (UAT)

After successful OAT (Operational Acceptance Testing), approval will be

given for Go-Live.

4.11. Training:
i. The bidder shall be responsible to provide offline mode training to relevant
user group (Minimum 30 representatives) of APDCL on site at APDCL HQ.
For the same, the bidder shall draw a training schedule (For a period of

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 34 of 115
 maximum 30 days) which shall be submitted to APDCL at the time of award
of contract.
ii. Bidder should provide training to APDCL staff on key implementation
aspects of the proposed solution and to efficiently executing their key
functionalities to obtain the desired goal for which the System was
implemented in the first place.
iii.The entire cost of training will be borne by the bidder.

4.12. Operation and Maintenance of the Service(s)

The service provider should provide handholding support for a period of 24 months.
This would include the following:

 Day-ahead, Intra-day, Real time and Long-term demand forecasting

 Day-ahead, Intra-day dispatch scheduling

 FMS support for a period of 2 years including free software updates

 Capacity building program for using the models for APDCL staff over the
entire contract period on need basis, with at least one in each quarter.

4.13. Handover of Software

I. The ownership all types of software modified/upgraded/developed along

with all design, architecture, code base, keys, user manual and other
related inputs along with all types of licenses for the purpose of contract
shall be transferred to APDCL at the end of contract period.

II. All the license(s)/ software required for the carrying out various activities
as per the scope of work shall be arranged by the Firm and the cost of such
license/ software shall be borne by the Finn for the period of contract.
However, the ownership all types of software
modified/upgraded/developed along with all design, architecture, code
base, keys, user manual and other related inputs along with all types of
licenses for the purpose of contract shall be transferred to APDCL at the
end of contract period.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 35 of 115
4.14. General Conditions

The cloud Data Centre shall have to comply with requirements of Tier III
category which applies to a concurrently maintainable site infrastructure with
redundant capacity components and multiple independent distribution paths
serving the critical environment. All IT equipment shall be dual powered. The
following general conditions will apply:
a) Only MeitY empanelled Cloud services should be used. Bidder needs to
provide valid document in this regard.
b) One of the most critical issues in the Cloud Service implementation is the
security of the data. It is the responsibility of the bidder to define the security
services that need to be implemented for their workloads depending on the
nature of the applications / data hosted on the cloud.
c) Bidder need to ensure that the Bidders facilities/services are compliant to
various security standards and should be verified by third party auditors.
d) BIDDERS should suitably address all the potential risks and issues in cloud
implementation including data security and privacy, increased complexity in
integration with existing environments, vendor lock-in, application portability
between different platforms, exit management / Transition-Out Services etc.
e) The Bidder shall be responsible for providing the cloud data centre services.
It shall be up to the Bidder, to identify the critical service agreements with the
concerned cloud data centre provider in order that the Bidder can meet and
sustain the SLA for this project.
f) All Services including data should be hosted in India.
g) Exit Management / Transition-Out Services -The responsibilities of the
BIDDERS during the Exit Management Period need to be agreed upon with the
Utility and they should assist the Utility in migrating the data etc.
h) The responsibilities of BIDDERS/MSP include migration of the data,
content, and any other assets to the new environment or on alternate cloud
service provider’s offerings and ensuring successful deployment and running of
the Utility’s Solution on the new infrastructure.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 36 of 115
 i) The BIDDERS should not be debarred/ blacklisted by any Government/ PSU
in India as on date of submission of the Bid.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL

Page 37 of 115
 10. Technical Specifications
10.1 Technical Specifications of Data Platform

10.1.1 DATA MANAGEMENT

The proposed solution should be an integrated solution containing Data Management, Data Quality,
1
Analytics, Alerting and Investigation, Monitoring and Visualization capabilities
The proposed solution should control access to applications, modules and functions based on user
2
roles and privileges.
The proposed solution should be configurable/customizable Commercial off The Shelf (COTS) as
3
per the requirements of this RfP to reduce implementation effort, time and cost.
The proposed solution should provide for services being started/stopped from the administrative
4
console
The proposed solution should control access to applications, modules and functions based on user
5
security
The proposed solution should support for encryption. The suite should provide for data, metadata,
6
data transmission between systems in encrypted format
The proposed solution should provide native access Industry leading RDBMS like ORACLE, SQL
7
Server, DB2, Teradata
The proposed solution should be rich in the set of in‐built transformations and functions that should
8 include predefined table and column‐level transformations and provide capability for reuse of
individual transformations
9 The proposed solution should support for surrogate key generation.
The proposed solution should have User interfaces for Data Profiling, Data Standardization, and
Clustering and Data Augmentation capabilities. In Data Profiling it should be able to conduct the
following analysis:
• Structure discoveries,
• Frequency distribution
• Pattern distribution
10 • Various Statistical Analysis
• Outliers Detection and Percentile reporting
• Relationship Discoveries
• Referential Integrity Analysis
• Redundant Data Analysis
• Enable business rule validation.
• Drill through analysis from graphical reports to transactional data
The proposed solution should support data quality measurement on an on‐going basis embedded into
11
batch and near‐time processes
The proposed solution should support data cleansing and de‐duplication, duplicate suspect
12 processing, house holding, with array of out‐of‐the‐ box standardization rules conform data to
corporate standards – or can build customized rules for special situations.
13 The proposed solution should provide fuzzy logic to induce tolerance during matching
14 The proposed solution should use the parsed data to provide flexible matching criteria
The proposed solution should enable to define rules for record and/or field selections during the
15
merging process
The proposed solution should have the ability to disable and enable nodes and submit flows in any
16
state.
The proposed solution should have capability for the process and the nodes selected in the
17 process flow diagram to generate its own code. The process flow should have the facility to
view and edit the code generated by the system.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL
Page 84 of 115
 The proposed solution should have a single interface for data integration & manipulation
18
such as tabulation, statistical analysis, econometric modeling and multi-dimensional analysis
The proposed solution should have the ability to easily capture and display performance
19 information such as real time, CPU time, memory use, input/output, and record count data,
with the ability to display this information as a table or as a graph.
The proposed solution shall be able to join data from multiple sources and support for
20
concurrent processing of multiple source data streams, without writing procedural code
The proposed solution shall facilitate data profiling based on dynamic, user defined
21 validation rules and support identification of user defined ‘events’ to trigger alerts (through
email reports) to authorities
22 The proposed solution shall support In‐memory data handling
The proposed solution shall support correction logic for Indian names, addresses, phone
23
numbers, identification ID and other identification proof documents and demographic details
The proposed solution should be able to create networks based on both transaction as well as
24
relationship-based data, and create a nodes and links among the entities specified
The proposed solution should be possible to identify common entity types which are super
25
hubs i.e. appear commonly in majority of transactions and treat them separately as per need
The Data Analytics OEM should be a Leader for recent three consecutive years in Gartner
26
Magic Quadrant for Data Integration and Data Quality Reports.

10.1.2 ADVANCED ANALYTICS

The proposed solution should contain a sophisticated and GUI based predictive modeling and
1
analytical workbench.
The proposed solution should enable identification of suspicious consumer profiles through a
2 judicious mix of anomaly detection, business rules, predictive modeling and network analytics
The proposed solution should help analysts to visualize complex network of relationships between
3 entities - such as people, places/ locations, things and events over time and across multiple
dimensions
The proposed solution should help analysts identify entity relationships that aren’t obvious, traverse
4
and query complex relationships, and uncover patterns and communities interactively
The proposed solution should have in-built modules for analysis of variance, multivariate analysis
and statistical algorithms to build prediction models such as Linear, Logistic, Non-Linear and
5 Quantile regression models, Generalized Linear models, Predictive partial least squares and Decision
trees
The proposed solution should provide in-built features and advanced techniques for the analyst to
6 detect rare events, anomalies and outliers and/or influence points to help determine, capture or
remove them from downstream analysis such as predictive models
The proposed solution should in-built modules for modern machine learning algorithms to build
7 predictive models - such as random forests, gradient boosting, artificial neural networks, support
vector machines and factorization machines
The proposed solution should provide a rich set of data mining algorithms that can be used for
8 classification, regression, clustering, detection of outliers and anomalies, feature extraction,
association analysis, and attribute ranking.
The proposed solution should support Clustering of entities that are either user Defined or
9 statistically chosen as best clusters, along with strategies for encoding class variables into the
analysis

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL
Page 85 of 115
 The proposed solution should have flexibility of high-performance imputation of missing values in
10 features using different statistical options such as mean, pseudo-median and with user-specified
values or with random value of non-missing values
The proposed solution should support automated algorithms which will help the end-users to run
11
multiple algorithms at a time and hence compare the results between them.
The proposed solution should enable automated model assessment and scoring, and generate the
12 associated model performance statistics and code for model scoring
The proposed solution should allow user to compare different predictive models on the basis of
13
different test statistics, and select the best model for deployment automatically
The proposed solution should provide multiple methods to visualize data mining models and provide
14
the user with sufficient levels of understanding and trust
The proposed solution should support processing, trend-analysis and modeling for forecast of
15 data‐points through exponential smoothing, missing data and outlier data on all data‐sets before trend
analysis / modelling
The proposed solution should support profile matching through user‐defined (configurable) business
16 rules through ad‐hoc querying across multiple fields of consumer‐wise information from in-house
and external agency data
The proposed solution should support Time Series and scenario (“What‐If”) analysis for dependent
17
variables.
The proposed solution should enable rule based / cluster analysis for profile grouping and profile
18
matching
The proposed solution should allow Analysts and Investigators to make use of a fraud intelligence
repository which gets populated containing information of performance of past models and scenarios,
19
to improve accuracy of current predictive models should be able to define risk based on different
levels such as relationships with entities, financial / non‐financial transactions & events etc.
The proposed solution should allow alerts to be generated whenever flagged entities or entities with
20
high risk rating and having financial/ non-financial transactions or some level of activity
The proposed solution should support detection of patterns so that criteria for various thresholds can
21
be reviewed and modified.
The proposed solution should provide in-built feature of detailing rule robustness through measures
22
of true positive, false positive and false negative as visual diagnostics
The proposed solution should provide an extensive list of prebuilt rule operators that would be
23
available to the analyst for detailed rule-model specification
The proposed solution shall support identification of common patterns / factors / profile
24
characteristics that could enable selection of criteria for selection of Business Audit cases
The proposed solution shall support analysis of variance, Multivariate analysis of variance and
25
repeated measurements and Linear and non‐linear mixed models.
The proposed solution shall provide a rich set of data mining algorithms that can be used for
26 classification, regression, clustering, detection of outliers and anomalies, feature extraction,
association analysis, and attribute ranking.
The proposed solution shall support detection of patterns from the transaction data set over a defined
27
time period for particular individuals / groups
The proposed solution shall support automated algorithms which will help the end-users to run
28
multiple algorithms at a time and hence compare the results between them.
The proposed solution shall support processing of data‐points for exponential smoothing, missing
29
data and outlier data on all data‐sets before trend analysis / modelling
The proposed solution shall support profile matching through user‐defined (configurable) business
30 rules through ad‐hoc querying across multiple fields of consumer‐wise information (across
application, usage, payments and other data)
The proposed solution shall allow identification of localities/ regions where high numbers of risky
31
consumer profiles are detected.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL
Page 86 of 115
 The proposed solution shall support analysis of voluminous data to identify patterns and determine
32
risk rating / payment reliability of a particular consumer
The proposed solution should be capable of payments matching ‐ i.e. reconciliation of payments by
33
consumers with that identified by the customer account statements.
The proposed solution should be capable of calculating network analytics and relationships among
34
consumers with a known risky / suspicious entity, thus enabling risk by association.
The proposed solution shall have capability to identify cases for consumers for whom the usage is
35
fluctuating at very fast rate. Such consumers are likely to be part of fraudulent activity.
The proposed solution shall have capability to estimate the liability of the consumer to pay or any
36
other sum payable and accordingly revise the risk rating.
The proposed solution shall support assessment of impact of an NTL on the revenue, workload on a
37
particular office location / region, impact on a particular group
The proposed solution should enable analysts to be able to carry out Collection optimization based on
38
projections of default as against resources to collect / follow up / investigate at ward / circle level
The proposed solution shall support detection of patterns so that criteria for various thresholds can be
39
reviewed and modified.

10.1.3 DATA WAREHOUSE

Product should be based on pure Massive Parallel Processing (MPP) Shared nothing architecture that
1 provides out-of-the-box fault tolerant and High available features
It should take full advantage of columnar database technology which provides automatic column
2
storage, column compression and indexing without additional DBA effort.
3 Support data encoding or compression on data storage to minimize and optimize the storage usage.
4 Support and provide native tools to handle performance monitoring and tuning
5 The product must be able to load and export data in parallel without any sequential choke points.
6 The product must be able to access external data sources in parallel.
The product must be based open standards to store and query data. it must support ANSI SQL
7 Standard and should NOT have any proprietary query languages of its own.
The product should be fully compliant with SQL-99 Standards, SQL2008 Standards and support
8
SQL2003 OLAP extensions
The product must provide native functionality to load and store XML/JSON data files within the
9 database and support search, query and update functionalities
The platform must be able to operate with variety of data models with no performance degrade (e.g.,
10
Star Schema, snowflake schema etc.)
11 The product must support both column encryption and network data encryption
The product must provide in-database machine learning and statistical analytics without dependency
12
on any external tool and language.
13 The product must support programming languages - Python, R for Machine learning purpose.
The product must support integration with procedure language like R, Java, C/C++, Perl, Python with
14
database
15 The product must support creating user defined functions and procedures natively using SQL.
16 Provide in-database SQL analytical functions like timeseries, sessionization
Support Industry standard API interface for user and application connectivity like
17
JDBC/ODBC/ADO.NET
DWH should have access to be connected by third-party reporting, query tools without any
18 limitations

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL
Page 87 of 115
 EDW system should support following aspect of performance:
§ Concurrency
§ Competing workloads
19 § Reporting, real time, OLAP, advanced analytics, etc.
§ Intraday data loads
§ Thousands of users
§ Ad hoc queries
EDW should have the capability to scale a data warehouse easily, efficiently, quickly, and cost-
effectively by adding units of computing power/disks/capacity, without undermining the increase in
20 linearity of performance and capacity per incremental resource keeping in mind both vertical and
horizontal scaling.
The system should have auditing and Logging Features for every action by user or system itself
21 without affecting performance of the system.
22 System Licensing should not have Limitation on number of Servers and CPU Cores
System Licensing should not have hardware infrastructure dependency and should be open to use on
23 any environment i.e. cloud, inhouse etc.

24 The analytical database should have inbuilt geospatial data analysis capability

Platform should support Docker and containers technology so that it can be launched standalone and
25
run as container on any Linux server
Platform should support infra-automation framework like Kubernetes so that high availability of
26 platform can be maintained with containers
Platform should provide query federation feature so that external data like parquet and ORC can be
27 queried from this platform and can be joined with internal table, without moving external data inside
platform.
The platform should provide inbuilt data exploration capabilities so that analyst and data scientists
28
can explore the data with native inbuilt functions and use need not to write code.
Platform should allow machine learning models to be exported and imported in standard format like
29 PMML. So that other platform developed/trained ML models can be used in this platform and vice-
versa.
Platform should support accessing and querying open data formats like Parquet and ORC on external
30
storage like NFS/S3/GCS/SAN etc without moving data into this platform
Platform should support querying other standard SQL databases without copying data from other
31 databases
Platform should be able to integrate with open-source Machine learning model and framework like
32 Tenser flow so that Deep learning forecasting can be done within data platform without moving data
out of the platform.
33 Platform should be complying with standard data Security Norms like FIPS 140-2 and equivalent
34 Platform should support Format Preserve Encryption FPE for data security purpose.

10.1.4 Public Cloud

The entire solution is envisaged to be hosted on Cloud Infrastructure and hence all
of the components of the proposed solution should be compatible with
virtualization. The solution should be hosted in Windows (latest version) or RedHat
Linux Operating System. Cloud provisioning and application hosting, including
operation and maintenance of associated application as per defined SLAs.
BIDDERS (Cloud Service Provider) must be empanelled under MeitY (Ministry of
RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL
Page 88 of 115
 Electronics and Information Technology) and STQC (Standardization Testing and
Quality Certification) Compliant during the entire contract period.
The successful bidder shall be entirely responsible for the services of the CSP for
maintaining required Uptime of the provisioned infrastructure for the data analytics
platform and the developed solution.
The solution has to be deployed in secured and dedicated cloud-based environment
in India. Bidder must ensure that all compute, storage, network, database and other
resources are hosted within India region only. Bidder also ensures that APDCL data
must reside only in India region.

10.2. VISUALIZATION AND REPORTING

Visualization and Reporting Specifications

The proposed solution should provide facility to summarize and present data using a variety of
1 highly customizable charts, including vertical and horizontal bar, pie, donut, sub‐grouped pie, star
and block charts, plots like scatter, line, area bubble, multiple axis and overlay plots
The proposed solution analytical reporting should enable administrators and solution designers to
2 drag and drop an entity’s fields onto a page template and Page Builder will choose the most
appropriate control to be used (based on the field type) and set the label and data source properties
The proposed solution analytical reporting should provide flexibility of displaying the dashboards
3
on third‐party tools such as MS‐Office applications
The proposed solution analytical reporting should allow generation of dashboard using ad‐hoc
4
queries by the user
5 The proposed solution analytical reporting should be web‐enabled.
The proposed solution analytical reporting should have capability to import and integrate local
6 text/csv/xls files with the data warehouse/ODS and be able to generate reports with no/minimal
intervention
The proposed solution analytical reporting should provide a user friendly; web based, drag and drop
7
interface for data preparation for data tables available in‐memory
The proposed solution analytical reporting should provide Geographical map views to provide a
8
quick understanding of geospatial data.
The proposed solution analytical reporting should allow 'On‐the‐fly' hierarchy creation for being
9
able to traverse to lowest information to undertake root cause analysis
The proposed solution analytical reporting should have the ability for Interactive report viewing for
information consumers using iPad and Android devices using a native application most popular
10
gestures and capabilities, including zoom, swipe, etc., to optimize ease of use and user engagement
while online or offline.
The proposed solution analytical reporting should provide capabilities to schedule jobs for data
11
updates and report refresh
The proposed solution analytical reporting should be capable of scaling up to distributed computing
12 environments if needed in order to leverage the parallel processing capabilities of multiple
machines
The proposed solution analytical reporting should be able to maintain Mobile device logging
13
history and also be able to backlist/whitelist devices
The proposed solution analytical reporting should allow to design reports and dashboards in a GUI
14
based environment with automatic refresh based on changes in underlying data sources
RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL
Page 89 of 115
 The proposed solution analytical reporting should have feature wherein Report can be drill down to
15
most granular level of detail as their access controls / profiles allow
The proposed solution analytical reporting should have feature wherein Parameters can be passed
16 among reports to retrieve details and investigate specific consumers / entities / location/ any other
segments.
The proposed solution analytical reporting should have feature wherein Based on need, users of the
17 reporting portal can execute stored procedures through the reporting interface and visualize the
reports
The proposed solution should have user interface with the capability to integrate with Web-Services
18
i.e. Should support sending and receiving web services
The proposed solution analytical reporting should provide for Conditional highlighting to provide
19
visual highlight information inside tables and graphs depending on conditions defined
The proposed solution analytical reporting should provide for Creation of filters to control data
20
displayed on a report
21 The proposed solution analytical reporting should provide for dynamic and cascading prompts
22 The proposed solution analytical reporting should allow to Add/view report comments and annotate
The proposed solution analytical reporting should have flash data visualization with multi-
23 dimensional view in a moving time-series
The proposed solution analytical reporting should have seamless integration (common technical
24 metadata) with Analytics (Statistical Analysis)
The proposed solution analytical reporting should Provide self-sufficient and easy access to
25 enterprise data sources on multiple platforms, operating systems and databases
The proposed solution analytical reporting should allow Analysis of results to be delivered in
various formats like HTML, RTF, text, and/or PDF form. Should provide the facility to save output
26 in tables or data as the same as the input format
The proposed solution analytical reporting should allow to Define report layout through wizards:
27 add tables, graphs, headers & footers
The proposed solution analytical reporting should support Graphical Query Builder for users to
28 visually access and manipulate their data without SQL expertise
The proposed solution analytical reporting should support Preview generated query code, check the
29 query syntax validity or copy the code for reuse in other applications
The proposed solution analytical reporting should allow collaboration within departments and users
by means of commenting and replying. Additional functionality of attaching supporting documents
30 to substantiate the comments should be possible
The proposed solution analytical reporting should support Visualizations to aid filter selection by:
· Filter all dashboard KPIs at once
31 · Data Brushing and contextual filtering between KPIs
The proposed solution analytical reporting should support Reports to create by the system which
should be dynamic and not static - i.e. - they should be interlinked, filterable and capable of drill
32 down analysis
The proposed solution analytical reporting should provide Auto charting. Based on data items
selected for analysis, the tool should automatically choose best visualization suited for
33 representation
The proposed solution Should support advanced analytics, statistical analysis, forecasting and
34 advanced aggregation
The proposed solution should provide facility to generate static or dynamic interactive visualization
35 charts and graphs
The proposed solution should support Visual depiction, highlighting and color coding based on
36 entity risk should be reflected in the network visualization/ other appropriate visualization
The proposed solution should support understanding of alert performance in terms of true positives
37 / false positives etc. should be assessable and measurable through reports

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL
Page 90 of 115
 10.3 ALERT AND INVESTIGATION FRAMEWORK
The proposed solution should incorporate two key aspects – one is a advance analytics solution
which provides dashboards and reporting at aggregated level for key officers and stakeholders, and
1 the other should be an investigative workbench which also enables operational stakeholders
(officers, enforcement department etc.) to take action on the alerts that emerge for a consumer,
assess the evidence around the same, and take a decision on further action on the alert.
The proposed solution should provide in-built features for Alert and Event Management with -
Governance, audit and compliance, Prioritized queuing model, Enrichment, Scenario-fired event
2
model, including scenario context, Manual alert creation and routing, Alert domains and Custom
disposition actions.
The proposed solution should provide a built-in functionality of alert based investigation and alert
3 exploration and triage - in which alerts are reviewed to determine the probability that they represent
suspicious behavior and are evaluated for their importance
The proposed solution should provide built in features to apply an appropriate disposition of the
alert - such as closing, suppressing, moving to another queue (such as high or low priority), linking
4
to a different object and sending the alert information to an external system after a decision is
reached about how to handle the alert
The proposed solution should provide an option to automatically disposition the alert when a new
5 alerting event arrives. For example, the alert may be automatically assigned to an investigative team
The proposed solution should enable to not just identify the entity against whom the alert was
6
created, but also related alerts, related entities and their interlinkages.
The proposed solution should provide facility to define rules and set threshold‐based alerts for the
7
same on the data used for query and analysis supported by solution
The proposed solution search feature should allow user to select/deselect entities of interest like
8
Name, ID, PAN, TIN etc. to narrow down the search results for enhanced understanding
The proposed solution shall have capability to generate scores for different entities based on the
9 formulae and select alerts with higher scores.
The proposed solution should have built-in capabilities to assign Alerts/Cases to different strategies
10
and queue for assignment from drop-down
11 The proposed solution should be web-enabled
The proposed solution should provide analytical capabilities such as Correlations, Regression using
12
predefined ontologies, Network Plot, Decision Trees, Scenario Analysis, Statistical Analysis
The proposed solution should have capability to enable officers to segregate cases separately for
13 defaulters and inconsistent usage
The proposed solution shall enable assessment of registration details for determining / modifying
14
risk profile rating and to detect fraudulent consumers
The proposed solution should have optimization capability to help in allocation of alerts for
15 different wards / circle members as per value at risk / capacity to investigate and probability of
recovery
The proposed solution shall provide facility to define rules and set threshold‐based alerts for the
16 same on the data used for query and analysis
The proposed solution should allow Networks to be visualized based on a temporal view – so that
17
the chronology of events are depicter through a time slider
The proposed solution should be able to activate / deactivate / reactivate / override specific alerts /
18 rules as per need at real time.
The proposed solution should be able to decide appropriate treatment for the alerts depending on
19
the rules under which they were flagged (route to investigator etc.) as per data event
The proposed solution should allow for Historical rule activity (alerts generated) and performance
20
(final disposition) to be identifiable which can be analyzed in order to improve rules
The proposed solution should be able to assign a unique case number to each item scored and
21
actioned by the Solution or out sorted for analyst review.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL
Page 91 of 115
 The proposed solution should have capability to visualize the network related to alerts or risk
22 networks. The networks as well as other alert reports should be capable of color coding to highlight
risky / high risk entities.
The proposed solution should allow investigator to look at the concise view of the network, as well
23 as, based on need, grow the network along a timeline or across entities to incorporate a larger level
of entities.
The proposed solution should enable analysts to be able to add comments and make notes in alerts /
24
investigations
25 The proposed solution should identify the risks associated with a given set of consumers
26 The proposed solution should route alerts to the appropriate investigator / person/group
The proposed solution should have Fraud Analytics solution to be integrated with Social Network
27
Analysis solution

10.4. ANALYTICS FOR FORECASTING AND DEMAND PLANNING

1 The Data Analytics OEM should be a Leader for recent three consecutive years in Gartner Magic
Quadrant for Data Science and Machine Learning Reports.
2 Aggregates transactional data into time series format and identifies and accounts for missing values.
Accumulates time-stamped data into any time interval (hours, weeks, months, etc.) for forecasting,
3 sophisticated modeling techniques.
Automatic outlier detection outliers with provision for selecting number of outliers, percentage of
4 outliers and sensitivity levels.
5 Trend analysis, Seasonality and intermittent series tests.
6 Missing value imputation using various interpolation and extrapolation methods.
Utilize comprehensive modeling techniques such as ARIMA, Neural Network, UCM, Generalized
7 Linear Models, and GAMS etc.
The load forecasting should be able to handle two stage modeling for residuals. The forecasted
8 residual should be able to be added back to the forecast load to generate the two-stage forecast load.
Special days and effects Management: Prebuilt Effects Management with diagnose process
workflow to move from Naïve load forecasting model to add effects and see improvements in a step
9 wise method with a holdout sample.
10 The effects that the solution should be configured to include -
a. Recency effect
b. Weekend effect
c. Holiday effect
Statistics should include prebuild measurements such as -
a. MAE, MAPE, and ME for annual energy, annual peak load, daily energy, daily peak load,
11 monthly energy, monthly peak load and hourly load
b. Ability to include macro-economic variables and also the cross effect of macro-economic
variables.
The tool should have Automatic generation of model selection lists and Choice of automation level
12 for all three forecasting steps: model selection, model parameter estimation and forecast generation.
The tool should have Hierarchical Load Forecasting and Temporal Load Forecasting. Exponential
smoothing models, ARIMA, ARIMAX, Unobserved Component Models. Utilize comprehensive
modeling techniques including ARIMA, ARIMAX, Neural Network, Unobserved Component
13 Models, Generalized Linear Models
The load forecasting should have pre-built two stage modeling techniques for residual forecasting
that adds back to the forecast load to generate the two-stage forecast load using Generalized Linear
14 Modelling, ARIMAX Modelling and Neural Network
15 Option for uploading a manual demand forecast shall also be provided by the Analytics solution.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL
Page 92 of 115
 10.5. USER ADMINSTRATION
The proposed solution should control access to applications, modules and functions based on
1
user roles and privileges.
The proposed solution should control access to applications, modules and functions based on
2 user security.
The system should have auditing and Logging Features for every action by user or system
3 itself without affecting performance of the system.
Proposed system shall support role-based access and interfaces for all types of users and
4 centralized administration.

10.6 Security
1 The proposed solution design should ensure the integrity and security of the data and
application.Integrated security measures regarding system, users and data sources.
2 Security: Adequate security features to ensure only authorized secure access.
The solution should employ the latest industry standard security tools and features required
3
to secure the platform, solution, access and protect from any unauthorized access.
4 Security audit for the solution may be performed by APDCL.

RfP for appointment of AI/ML Powered Business Intelligence Analytics Solution Provider for APDCL
Page 93 of 115