Rising living costs are intensifying food insecurity, heightening the need to reduce avoidable food waste (FW). Quick Service Restaurants (QSRs), built around fast and affordable meals, contribute substantially to FW, with the hospitality sector responsible for about 16% of Australia’s total FW (1). Limited comparable data and a lack of FW management tools tailored to small- and medium-sized QSRs make routine, self-managed monitoring and reduction difficult without researcher support (2,3). This study applies Statistical Process Control (SPC) to identify FW trends and patterns and, together with self-reported and observed foodservice practices, develops an adaptive sampling schedule that enables QSRs to monitor FW less frequently while maintaining comparable accuracy. Audit data for each outlet were analysed using SPC to quantify volatility in FW generation and to detect significant trends and outliers across the audit period, incorporating both continuous and ordinal variables collected during the audit period. Four QSRs in the City of Adelaide, South Australia (n=4) completed baseline FW audits over 15-18 days, quantifying pre-consumer FW across spoilage, preparation waste (edible and inedible) and overproduction using a validated direct weighing method (3). Spent coffee grounds were collected separately where relevant. Kitchen and menu practices were observed, and semi-structured interviews with owners/managers verified operational processes. SPC analysis was used to assess volatility in FW generation. Adaptive sampling frequencies were derived from SPC charts and observed/self-reported practices, with reduced-frequency audits conducted over an equivalent period. FW data were normalised and compared between audit periods for each QSR. SPC charts for all FW categories, including spoilage, edible and inedible preparation waste, spent coffee grounds and overproduction, revealed predominantly common-cause variations within and between operating days, alongside distinct category-specific patterns across QSRs. Hotspots varied by menu style and kitchen practices, with spent coffee grounds emerging as a major hotspot for QSRs offering coffee. Normalised FW data compared favourably between both baseline and reduced-frequency audit periods for each QSR. Findings indicate that an adaptive sampling approach can reliably estimate categorised FW with comparable accuracy to daily baseline monitoring, enabling QSRS to identify reduction opportunities such as improved storage, inventory optimisation, refined menu design and portion control. A practical, self-managed FW monitoring method supports QSRs in implementing and sustaining their own reduction strategies, upgrading practices, educating staff, and enhancing business sustainability while contributing to national FW reduction targets. By conserving nutrients and resources, this approach also supports food security and reduces preventable losses that limit nutritional availability for at-risk communities.
Keywords: pre-consumer food waste; sustainable food systems; statistical process control; adaptive sampling