Circular Economy Software Is Failing Where It Matters Most: The Design Table

The manufacturing industry is one of the great engines of modern civilisation, generating more than US$16.82 trillion in value in 2024 and supporting hundreds of millions of jobs worldwide. It is also one of the great engines of environmental damage. The sector consumes roughly 54 percent of the energy supplied globally and accounts for a substantial share of greenhouse gas emissions, while humanity’s appetite for raw materials has tripled in half a century, climbing from 30 billion tonnes of extracted material in 1970 to more than 90 billion tonnes in 2020. As pressure mounts on finite resources and waste streams swell, researchers and companies alike have turned to the circular economy, a framework that keeps materials and products in use through reuse, repair, remanufacturing, refurbishment and recycling rather than letting them flow linearly from factory to landfill.

But a new systematic review of the scientific literature suggests that the digital tools designed to make manufacturing circular are themselves stuck in a linear mindset. The study, published in the Journal of Industrial Ecology by Themiya S. Kuruppuge and colleagues at the University of Exeter’s Exeter Digital Enterprise Lab, analysed 231 papers drawn from an initial pool of 614 documents indexed in the Scopus database between 2000 and 2025. Using bibliometric mapping with VOSviewer and Biblioshiny alongside qualitative content analysis, the team set out to answer five questions about how decision support systems, or DSS, are being used to embed circular economy principles across manufacturing, from the drawing board to the reverse logistics network.

The headline finding is stark fragmentation. Only 5 percent of the reviewed studies integrated decisions across all three levels of the manufacturing operation simultaneously: the product level, where designers specify materials, geometries and functional requirements; the process level, where engineers control energy use, waste and emissions along the production line; and the system level, where supply chain managers design forward and reverse logistics networks, choose facility locations and coordinate suppliers, manufacturers, distributors and customers. More than half of the studies, 61 percent, addressed just a single level, with 46 percent focused on product-level questions such as material selection and design for disassembly, and 39 percent on system-level network design. The remaining third of the frameworks linked two levels, but full integration remains vanishingly rare.

The review also uncovered a powerful temporal bias. Forty-five percent of the decision support frameworks concentrated on end-of-life decisions, such as recovery strategy selection, disassembly planning and recycling process optimisation, while only 8 percent addressed the manufacturing phase. This end-of-life dominance reflects the field’s roots in waste management and recycling, and the immediate regulatory pressures manufacturers face to collect and process returned products. Yet the review points to a critical mismatch: studies cited in the paper suggest that around 80 percent of a product’s environmental impact and circularity potential is fixed during the initial design stage, precisely when designers have the least access to lifecycle information. Products designed without consideration of recovery requirements arrive at processing facilities constrained by upstream decisions, limiting the achievable level of circularity no matter how sophisticated the recycling operations downstream may be.

The authors frame this as an information-decision timing mismatch, and it is arguably the central technical challenge of the field. Designers make material selections and architectural choices during conceptual phases with sparse lifecycle data, while the information needed to evaluate those choices only becomes available after products have been produced, distributed, used and recovered. Traditional life cycle assessment, the workhorse method for quantifying environmental impacts across a product’s lifespan, is frequently deployed to measure the impacts of finalised designs rather than to optimise them early on. Limited information and high uncertainty during early design phases hinder the use of LCA exactly where it would deliver the greatest leverage, and many of the frameworks in the literature remain conceptual, struggling to reach industrial practice because of data unavailability, complexity and high collection costs.

On the assessment side, the review found that environmental and economic tools dominate. Life Cycle Assessment and Life Cycle Costing frequently appear together, quantifying ecological performance and tracking costs across lifecycles, which matters particularly for remanufacturing and product-service systems where upfront investments differ from traditional manufacturing but generate returns over extended periods. Techno-economic analysis extends this by testing whether recovery technologies can meet technical specifications while remaining economically competitive. Material Flow Analysis identifies where materials accumulate, leak from circular loops or get downcycled into lower-value applications, while circularity metrics such as the Material Circularity Indicator quantify the proportion of materials from recycled or reused sources flowing into future use cycles. Mathematical optimisation models, including linear and mixed-integer programming and multi-objective optimisation, are then used to identify optimal solutions from within these frameworks.

Digitalisation is the fastest-moving frontier. The bibliometric analysis revealed three distinct evolutionary phases: early research from 2000 to 2018 concentrated on closed-loop supply chains, reverse logistics and product recovery; a middle phase shifted attention to assessment methodologies and eco-design approaches such as design for recycling and design for disassembly; and the most recent phase, from 2023 to 2025, is defined by digital technologies, Industry 4.0 and digitalisation. The tools now on the table include Internet of Things sensors that generate data on actual use patterns and remaining useful life, digital twins that simulate lifecycle performance before physical products exist, machine learning models that predict future return characteristics, and blockchain systems that maintain product history across ownership transfers. These technologies do not eliminate the timing mismatch, but they accelerate information availability and allow decisions to adapt as product conditions change, providing the connectivity, traceability and data transparency that circular operations demand.

Yet the review is clear that technology alone will not close the gaps. Information must reach decision-makers in forms they can actually apply: manufacturing engineers need process specifications rather than raw LCA data, and supply chain planners need return volume projections formatted for logistics optimisation tools. There is also a striking blind spot around business model innovation. The circular economy fundamentally requires new ways of creating value, such as product-as-a-service, leasing, pay-per-use and sharing schemes, which change the decisions that must be made and the information needed to support them. When manufacturers retain ownership of their products, they gain direct incentives to design for durability, repairability and multiple use cycles, aligning business interests with circular principles. But these models introduce complex pricing questions, capacity planning that must balance new production with refurbishment, and new stakeholders including third-party service providers, financing institutions and insurers. Recent industry surveys cited in the paper report that more than 70 percent of manufacturing leaders expect circular business solutions to raise revenue by 2027, and 65 percent believe such solutions will enhance operational resilience.

The geography of the research field is itself uneven. Some 56 percent of the reviewed papers originate from European countries, a pattern the authors attribute to strong policy frameworks including the EU Circular Economy Action Plan of 2015 and its 2020 update under the European Green Deal, which mandate circularity considerations in product design, production and waste management and introduced instruments such as digital product passports and right-to-repair provisions. Less than 5 percent of studies came from South America, Africa and Oceania, suggesting that much of the global manufacturing base is developing circular decision support outside the mainstream research conversation. The authors acknowledge their own limitations as well: the review drew only on Scopus, included only English-language documents, and excluded the construction sector, so regional innovations and non-English literature may be underrepresented.

The way forward, the authors argue, is an integrated decision support system that converges system-level thinking with a circular economy framework and digital capabilities, creating an early-stage tool that provides feedback during design phases rather than after the fact. In their proposed vision, product, process and supply chain decisions operate within a shared design space, influencing one another rather than functioning in isolation, with digitalisation as the enabling layer that makes cross-level, early-stage support practically achievable. The deeper message is a reframing of sustainability itself: not as a constraint checked after designs are finished, but as a design goal optimised alongside functionality, cost and quality. Achieving that will require interdisciplinary collaboration across engineering, operations and business strategy, and a willingness to treat the circular economy not as an end-of-pipe fix but as the organising principle of how products are conceived in the first place.

Subject of Research: Decision support systems for circular economy implementation in manufacturing supply chains

Article Title: Decision support systems to facilitate circular economy at the supply chain level: a systematic literature review

Article References: Kuruppuge, T. S., Kulatunga, A. K., Luis, M., & Sucala, V. I. (2026). Decision support systems to facilitate circular economy at the supply chain level: a systematic literature review. Journal of Industrial Ecology, 30(4), 2141-2157. https://doi.org/10.1007/s44498-026-00145-6

Image Credits: AI Generated

DOI: 10.1007/s44498-026-00145-6

Keywords: circular economy, decision support systems, sustainable manufacturing, supply chain management, life cycle assessment, Industry 4.0, digital twins, reverse logistics, eco-design, business model innovation, systematic literature review, industrial ecology