Vegetation management: Emerging tech to unlock efficiency
March 15, 2021
March 15, 2021
Vegetation Management poses operational and maintenance challenges for utilities and there is considerable internal and external pressure to increase efficiency in this area of the business. It is typically managed by third party suppliers, and requires crews to manually survey and identify vegetation that needs pruning, trimming and removal. The task is critically important for public and environmental safety.
In this blog, I explore how advanced analytics and technology can be leveraged to unlock efficiency, productivity, improve safety and transform how utilities manage vegetation management work.
Trees are the leading cause of electric service interruptions, especially during major weather events (e.g., ice storms, hurricanes) and can create public and employee safety hazards, so improving efficiency requires a thoughtful and planful approach to program optimization.
While it is difficult to replace labor that performs line clearance, this work can be optimized using emerging technology, tools and advanced analytic approaches, without reducing reliability, safety and customer satisfaction
Here, we focus on three specific aspects of the program.
Some utilities trim their circuits on a fixed cycle (usually 4 or 5 years) or use outage data from the previous year to determine what the next year’s circuit trim list should be. However, neither one of these approaches takes into account the longer-term experience and view of the reliability associated with the different circuits. Using advanced analytics and tools, leading utilities have developed predictive models to create predictive failure and trimming cost curves to optimize the trim cycle at the individual circuit level.
Tree Trimming Model (TTM) is one example of such a tool that can be used to model 10 years of reliability related to vegetation management based on different cycles, budgets and targets. It has been used to support utilities discussions and agreements with regulators to secure sufficient funding for the vegetation management preventive maintenance program. In an actual scenario, where a utility was on a fixed 5-year cycle (i.e., trimmed each circuit on a 5 year cycle), TTM was able to take the same spend over a ten-year period, optimize it and improve reliability by 10% over that period of time
In order to increase the productivity of the VM crews, utilities typically send planners to “walk the circuits” that are identified for trimming and determine what trees have to be trimmed, which ones need to be removed and which customers need to be notified. During the COVID pandemic, these in-person discussions with property owners to get their approval for removal became much less common. As a result, there was an opportunity to rethink the entire process and figure out how to leverage image analytics (e.g., satellite, drone) to identify the work and then submitting approval requests to property owners electronically (e.g., e-mail, text).
While there are challenges in using this type of advanced analytics in the areas where there is overhang and satellite imaging cannot certify the distances, it can still be used to identify 80% of the work that needs to be done by applying machine learning (ML) and artificial intelligence (AI) algorithms to image analytics.
These technologies are already available and there is an opportunity to deploy them and leverage the experts (i.e., Foresters) to train these algorithms and build confidence that they are working properly over the next couple of years. At that point, the results from this analysis can be used to further refine the priorities, optimize the crew assignments and minimize the cost of acquiring property owner approval.
The experienced resources and experts should be used to deal with any escalations from the field and manage more complex issues, while the technology is used to build 80% of the plan.
As we mentioned previously, most of the field work related to the VM program is conducted by third party suppliers. In order to ensure that the circuits which were trimmed have adequate clearances, utility foresters conduct either 100% or statistical sample QA/QC reviews of the work.
Sometimes these field inspections are done soon after the work is completed and other times a few weeks or months later. When the inspectors find an issue, they refer it back to the contractor who then has to roll another bucket truck to make the correction, which one way or another adds costs.
With the proliferation of the technology and image analytics, the QA/QC process can be almost fully automated.
Imagine the future, where the trimmer can take a picture of the span that have just completed using an app that automatically submits that image for analysis against a database using ML and AI algorithms.
The trimmer then gets immediate feedback that identifies where there may be a lack of clearance, so that the crew can move the bucket truck 15 yards and fix the issue right away.
This way, they would eliminate the need for rolling another truck and minimizing the need to conduct field reviews of the contractor work. Finally, it provides a clear record of the work (from the images) and the audit path for the invoices.
Contact me to find out how Accenture’s vegetation management solution is helping our clients on their journey towards a more digital, resilient, and distributed grid.