1. Energy, Utilities and Chemicals the way we see it
Conducting a GIS Data
Refresh—The Foundation
of Your Smart Grid Program
GIS refresh strategies, processes and considerations
Introduction management, engineering design and
After worldwide spending on Smart construction and to support advanced
Grid initiatives in 2008 exceeded applications such as outage management,
$12 billion—a total expected to GIS has been around most electric
surpass $33 billion within the next utilities for many years.
five years—utilities are intensely
focused on maximizing their Smart As they begin to implement Smart Grid
Grid investments. To that end, initiatives, utilities are increasingly
companies are reviewing the state of realizing that the accuracy of the data
their existing infrastructure systems. stored in their GIS is essential to their
To successfully meet the future business projects’ success. As a result, companies
expectations of Smart Grid, the are concentrating on two GIS issues that
availability of complete and accurate significantly impact Smart Grid
grid data is essential. Therefore, the implementations: how the initial GIS
Geographic Information System (GIS) was implemented (how data was
is a critical component. modeled) and the quality
Installed for a number of reasons
including paper mapping
displacement, asset

2. of the electric network data contained in tracing it to the end of each branch. are visible, reasonably easy to access and
the GIS. Utilities that implemented GIS During the walk-down, the operator fairly easy to understand. Operators need
as an asset model frequently discover validates all current GIS data and only to start at the substation then follow
that the electric network data may not be collects any missing data. In addition the wires, noting any electrical device or
electrically connected creating complete to correcting the current GIS data change. They do not need special skills
elements, if Smart Grid initiatives or experience that cannot be provided
electric feeders as required by Smart
require data that is not currently through simple training. They can be
Grid. The other issue of data quality is maintained in the GIS, the operator outside contractors or unassigned
obvious -- with incorrect network feeder must update the GIS data model and internal staff. Crew size can be one-man
data, the Smart Grid is unable to collect the additional data during the or two-person depending on local utility
correctly analyze and operate the walk-down. This is usually done by requirements and desires. If the feeder
electric network. printing paper feeder maps from the can be driven, sometimes the increased
GIS and marking up the maps with cost of two-man crews can be offset by
Solving both of these issues requires discrepancies as the operator walks the increased productivity. An allocation
refreshing and correcting both the the feeder. should be provided to cover overhead
model and the data contained in the and transportation costs.
GIS. However, as companies quickly The data to be captured or updated
learn, conducting a GIS data refresh is depends upon the utility’s data Factors that affect productivity
more complex than performing an requirements. However, it typically include the type of electrical
construction, the number of feeders
per right-of-way, feeder accessibility
(backyard, alley, etc.), number of
devices per feeder, number of
customers per feeder, the amount of
new data to be collected and most
importantly, the number of GIS data
errors detected that must be
documented. An overhead walk-down
usually has minimum safety concerns
since the walk-down staff is not
required to make contact with
electrical equipment.
Underground
A walk-down of underground
construction is much more complex
and expensive. Unlike overhead,
initial GIS data loading. For most includes electrical connectivity, circuit underground construction requires
utilities, the initial data loading was phasing, electrical asset information, sizes, inspection teams to open enclosures,
easier to estimate because all data had ratings, and conductor configurations. man-holes and vaults. This increases
to be validated and input into the the complexity, safety concerns and
system. With a GIS data refresh, all the Two major factors impact walk-down crew skill requirements. These issues
data must be reviewed and validated costs: the labor rates of the walk-down usually result in the need for
(typically called a “walk-down”) but field operators and their productivity. supplementary crews and equipment
only new data and errors must be While the labor rates can usually be support to meet safely requirements.
updated ("data posting"). Because many easily determined, field crew productivity The result is increased costs and
factors can affect the costs of a refresh can be impacted by a number of factors reduced productivity. An allocation
effort, the purpose of this paper is to which vary between overhead and should be provided to cover overhead
review and discuss GIS refresh underground construction. We will look and transportation costs.
strategies, processes and considerations. at each type of walk-down separately
because they vary significantly. Factors that affect productivity are
Feeder Walk-Down basically the same as for overhead with
The feeder walk-down is the process of Overhead the added complexity of exposing the
an operator physically walking each The easiest and least costly walk-down is underground network for inspection.
feeder starting at the substation and of overhead construction. These feeders Again, these factors include the type of
2

3. Energy, Utilities and Chemicals the way we see it
electrical construction, the number of change, which is dependent upon the ■ Neighborhood characteristics
feeders per right-of-way, feeder GIS system tools available and the - Urban or rural
accessibility (vault, man-hole, etc.), business process of the utility, you can
- Front lot or back lot construction
number of devices per feeder, number of determine the cost per update and per
(can you drive or must you walk?)
customers per feeder, the amount of new feeder. You must also provide for the
data to be collected and most importantly, QA/QC processes, which will add - Ease of access to facilities
the number of GIS data errors detected additional time and costs per feeder. (high, medium or low)
that must be documented. - Tree coverage
First Steps – Forecasting Model After an initial analysis of the feeder data
Data Posting The first steps in any GIS refresh are to and characteristics has been performed,
Following the walk-down and the estimate the cost of the data refresh, a sample set of representative feeders
documentation of the GIS data to be forecast the expected benefits and should be selected for a pilot walk-down.
updated, the next major step in a GIS develop the business case. To estimate The goals of the pilot are to capture
refresh is the posting of the documented the effort and cost of a GIS refresh, it is metrics data on crew performance and
update data to the GIS. This process is necessary to build a cost forecast model. the time required to walk-down the
time-consuming but relatively This is accomplished by first performing feeder, as well as to capture a sample
straightforward. Many companies an analysis of the GIS data for the feeders of update metrics.
outsource the data-posting process, to be refreshed. Any new data
usually to an off-shore firm to benefit
from the reduced labor costs.
The first step is to transfer the data
updates to the off-shore vendor. If the
data updates were captured on paper
maps, there are two options to
transferring the data: the maps can be
copied and mailed or they can be
scanned and electronically transferred.
The data-posting vendor’s job and cost
is determined by three factors: the
number of updates to be posted, the
productivity of the GIS software tools
and the business processes imposed by
the utility. It is assumed that the vendor’s
staff is skilled and trained on the
appropriate GIS system. requirements must be accounted for and Walk-down Forecast
included in the GIS model. Following the pilot walk-down, the
The vendor normally uses the following team should conduct a statistical
factors to estimate the total cost for The initial analysis should capture as correlation analysis to determine
data posting: the number of feeders, much data as possible about the which feeder characteristic variables
the number of conductor segments feeders including: are significant and contribute most to
(not conductor miles), the number of ■ Facility counts (all major facility types) estimating the cost and effort required
pieces of electrical equipment, the to walk-down all the feeders selected
■ Distance measurements
number of customers per feeder, the for refresh. A separate analysis will
(miles of conductor)
number of breaks or jumpers, and the need to be conducted for overhead
■ Conductor segment counts
estimate of the percentage of errors that and underground feeders since the
will require posting. ■ Construction characteristics
characteristics and crew requirements
- Overhead vs. underground are so different. The correlation analysis
By totaling the number of facilities on a can be conducted using one of many
- Three-phase or single phase
feeder and assuming that 30 percent of different tools available on the market
feeder data requires updating, you can - Multi-circuit right of way sharing
including Microsoft Excel. The results
calculate the number of updates. If you - Vertical vs. horizontal on overhead of the statistical correlation analysis will
assume three to five minutes per
Conducting a GIS Data Refresh—The Foundation of Your Smart Grid Program 3

4. www.us.capgemini.com/energy
enable you to construct a forecast Summary Remember, one reason for the GIS
model using the significant feeder For most utilities, the key source of both data refresh is poor data quality.
characteristics to forecast the time and grid asset and feeder connectivity is their Assuming you invested in good data
cost to walk-down all the feeders with GIS. Analysis of systems across North when the GIS system was initially
similar characteristics. Since the pilot America and throughout the world has installed and the data quality has been
sample size is limited, the forecast shown that while most utilities have deteriorating, you will need to protect
model can be improved as more feeder invested significant effort and cost in your refresh investment with
metrics are captured. creating their GIS, their data quality has improved business processes. The
been in steady decline due to lack of bottom line? The walk-down is only
Posting Forecast business emphasis and ineffective as good as the business processes that
The approach for developing a supporting business processes. In keep the data maintained.
forecasting model for data posting is addition to data deterioration, the Smart
similar to the walk-down model but the Grid may require additional data not
variables are different. The two factors currently modeled in the GIS. This new
that most impact the posting are the data capture could be significant if the
number of data updates required and the current GIS data model is inadequate.
productivity of the operator doing the
posting. The number of updates required In fact, depending upon the state of data
depends on the number of errors in the corruption and the number of new data
existing GIS feeder data and the amount elements, the GIS data refresh project
of new data that must be posted. The can be a significant undertaking. That’s
pilot walk-down should give you some why the utility should assemble an
idea of the error rates as a percentage of experienced project team, with
facilities on a feeder. The amount of new experience in GIS and GIS data capture.
data can be determined from the data
requirements and from the pilot walk-
down metrics. This data then can be
statistically analyzed to develop a
forecasting model. The operator
productivity is based on several factors.
The most significant are the GIS system About Capgemini and the
being used and the software tools Collaborative Business Experience
®
®
provided with the system. In addition,
the business process imposed by the Capgemini, one of the aims to get the right balance of the best
utility can have significant impact. The world's foremost providers talent from multiple locations, working as
business process can range from simple of consulting, technology and outsourcing one team to create and deliver the
editing (fastest) to a complex asset- services, enables its clients to transform optimum solution for clients. Present in
management work order process (much and perform through technologies. more than 30 countries, Capgemini
Capgemini provides its clients with insights reported 2008 global revenues of
slower). And finally, the individual skill
and capabilities that boost their freedom to approximately $12.74 billion and employs
and talent of the operator comes into achieve superior results through a unique over 90,000 people worldwide.
play as well. However, with this factor way of working, the Collaborative Business
the operator tends to increase in skill as Experience. The Group relies on its global More information is available at
the job progresses. As always, the delivery model called Rightshore®, which www.us.capgemini.com/energy.
quality-control process must be added
into the overall posting process.
The pilot sample will allow you or the
vendor to capture metrics on posting
EUC_SGGISDTPOV_051209_025
productivity and cost. This data can then Gord Reynolds
be used to develop a forecasting model. Practice Leader
If the utility used an off-shore data Smart Energy Services
conversion vendor, the vendor will have gord.reynolds@capgemini.com
a great deal of experience in forecasting +1 416.732.2200
the effort and cost.
Copyright © 2009 Capgemini. All rights reserved.