California Jurisdictions Usher a New Era of Progress Towards the Circular Economy

SEATTLE (Waste Advantage): When it comes to the implementation of SB 1383, California is not messing around. By 2025, the Short-Lived Climate Pollutant Law (SLCP) aims to achieve a 75 percent reduction in the level of organic waste disposal from the 2014 level.

According to CalRecycle, 480, or 78 percent, of California jurisdictions have reported that they have implemented residential organic waste collection. That is 8 percent more of the 617 California jurisdictions than what was reported in 2023.

Nineteen more jurisdictions plan to implement residential food waste collection in the second half of 2024. That means more than 80 percent of the state is nearly ready to move on to the next phases—Monitoring and Enforcement.

As expected, and documented in Zabble’s 2023 market research report funded by the US EPA’s SBIR Phase I program, there is an increased demand to safeguard the regulatory requirements of SB 1383 by implementing an effective monitoring and enforcement policy.

As jurisdictions start to focus more on monitoring and enforcement in 2024 and beyond, there are three themes that we have identified to help jurisdictions think about their strategy to go above and beyond compliance: technology, benchmarking, and targeted outreach.

#1: Technology
Choosing the right technology is vital for the success of a good monitoring program. Artificial Intelligence (AI) applied through robotics, sensors, and handheld mobile applications has made it possible to identify material composition at very granular levels at MRFs, collection trucks, and carts. Needless to say, the recent explosive growth in computer vision AI and Generative AI only underscores the fact that AI is going to continue playing a role in helping waste and sustainability professionals become more efficient in standardizing data collection. This will lead to jurisdictions gaining the proper insights to accelerate towards their goals around SB 1383 compliance, enforcement, and reporting.

As prescribed by the SB 1383 regulations, there are two ways to satisfy the container contamination minimization requirement: (1) Waste Evaluations and (2) Route Reviews.

Waste Evaluations
Waste Evaluations should take place at a permitted solid waste facility, such as a material recovery facility (MRF) or transfer station. It is currently conducted by jurisdictions, haulers, and waste consultants who manually sample different areas of the jurisdiction and sort them at the facility to identify prohibited contaminants and their weight. AI can help automate the process of identifying the prohibited contaminants and determining the ratio of prohibited contaminants to total sample weight without the need for sorting. The regulations suggest the minimum number of samples required based on the number of generators and hauler routes.

Route Reviews
Route Reviews are an alternative method of inspecting individual containers for prohibited contaminants such that all hauler routes are being reviewed annually. Currently, The two most prevalent ways for conducting route reviews are through truck cameras and lid flips. While the upfront cost for truck camera-based AI solutions is high, it is mitigated with more consistent route coverage throughout the year and fewer staffing requirements. People-powered lid flips avoid these high upfront costs, as the timing and coverage can be controlled. Lid flips also provide immediate feedback to the generator in the form of cart tags, whereas the use of truck cameras requires later follow-up with a mailer or other form of outreach.

Regardless of the method jurisdictions select, the use of technology and AI in the aforementioned areas can significantly increase the efficiency of these activities and reduce safety risks associated with handling sharp or hazardous material.

#2: Benchmarking
Historical data can be a great indicator of trends and a tool for informing key decisions. Benchmarking against past waste evaluations and route reviews help to optimize results each time they are conducted. The most important aspect of benchmarking is identifying relevant metrics to drive positive results.

Operational Benchmarks
Operational Benchmarks are key milestones to determine how well a route review has been conducted. Some of the metrics related to route reviews are:
• The duration to completely inspect a route
• The number of stops or addresses inspected per route
• The number of containers inspected per route
• The number of routes inspected per day
• The mean (or median) number of stops or addresses inspected in an hour

AI Benchmarks
Similarly, there are AI Benchmarks that can help us understand how well different AI models perform under different circumstances and environments. Computer vision AI or Generative AI can be defined by its performance related to bin fullness classification and object detection for contamination identification. Examples of metrics that capture their performance are:

Precision: The ability of a model to identify only the relevant objects. It is the percentage of correct positive predictions.
Recall: The ability of a model to find all the relevant objects. A high recall score indicates that the model effectively identifies most of the relevant objects in the data.

AI benchmarks in the waste industry, if published, tend to be opaque. Asking for AI benchmarks and an explanation of how they were measured can help shed light on the effectiveness of an AI-driven solution.

Courtesy: www.wasteadvantage.com