Exploring the effect of fashion obsolescence time on total garment discards
Bibliographic Data
| ID | 21652171 |
|---|---|
| Authors | Jim Duggan (0000-0001-8554-2594, School of Computer Science University of Galway Galway Ireland, corresponding author) |
| Year | 2024 |
| Volume | 41 |
| Issue | 6 |
| Pages | 871-875 |
| Publication date | 2024-11-01 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Systems Research and Behavioral Science (JOURNAL) |
| Journal identifiers | ISSN: 1092-7026 • E-ISSN: 1099-1743 |
| Publisher | Wiley (PUBLISHER • GB) |
| DOI | 10.1002/sres.3103 |
| OpenAlex | W4404883434 |
| Language | EN |
| Citations received | 1 |
| References cited | 6 |
In their paper exploring fast fashion principles and mass production within globalised supply chains (Martinez-Jaramillo & Tilebein, 2024), the authors address a wicked problem, namely, a class of problem that is ill-formulated, with many clients and decision makers with conflicting values (Churchman, 1967; Termeer et al., 2019). An aspect of this complexity can be observed through the multiple influences of modern technology on their model. For example, they present digital textile microfactories as a promising technological solution to reduce waste, yet at the same time they note that social media technology is a key enabler of trend chasing within the artificial demand reinforcing loop. Their work highlights the utility of qualitative modelling (de Gooyert et al., 2024; Lane et al., 2016) with causal loop diagrams (CLDs) that help explore dynamic patterns, propose solutions and identify possible leverage points, as well as providing a basis for developing quantitative models. Motivated by their work, we now present (1) an extension to their causal loop model in terms of an additional balancing feedback loop and (2) an initial exploratory stock and flow model, with sensitivity analysis, focusing on the model variable fashion obsolescence time. For the extension of the causal loop model we focus on loop R2, artificial demand. Our approach is to consider one of the loop's variables that contributes to the reinforcing behaviour, fashion obsolescence time. In the original model, this moves in the opposite direction to total demand for garments. We propose a balancing factor to this which is that as total demand for garments increases, so too does the demand for higher quality garments. As this demand is fulfilled, this leads to an increase in fashion obsolescence time, and so the overall impact of this change is that a new loop is emerging, a balancing loop, which we label ‘quality not quantity’ (Figure 1). With the focus on the fashion obsolescence time, we can now address this variable within a quantitative stock and flow model. The translation from a CLD to a stock and flow model, for simplicity, proposes a one-stock model for garments (garments in circulation), with garments produced as an inflow and garments discarded as the solitary outflow. The quantitative model is shown in Figure 2, and the completed Stella implementation (and R sensitivity code) can be accessed online.1 We run the model for two scenarios to highlight its behaviour through the stock total discards. For the first run, we fix the fashion obsolescence rate at 1.0, and this models fast fashion where people are discarding and replacing garments at a high rate. The second run sets the obsolescence rate to 0.5 (2 months), and the overall discards are reduced by 50%. While the results of these two runs would appear obvious, to explore this further, and gain additional insights into the relationship between total discards and fashion obsolescence Rate, we utilise sensitivity analysis (Figure 3). The simulation model is run 2500 times, once for each combination of the two parameters. The maximum for total discards is recorded for each individual run (the final value of the stock), as this is a useful overall measure of how successful the policy was. A scatter plot showing a random sample of 250 results is displayed in Figure 4, where each point size is proportional to the maximum size of total discards for that run. As expected, the plots show that total discards is smaller in the lower section of the graph, with longer lifetimes for garments and quick response times to the overall goal. The least desirable solutions are in the area where the desired fashion obsolescence rate is highest, as that corresponds to a shorter delay time for the garments in circulation. We can gain further insights into these relationships by exploring a contour plot (exploring a 3D surface in 2D), as this shows the parameter combinations that generate similar overall results. For example, in Figure 5, the navy shading indicates all those solutions that are in the range of 0–106, while the yellow band captures those results in the range of 9 × 106 and 107. While this is a simple model, nonetheless, the process highlights the potential value of initially building a simple simulation model that is informed by the causal loop structures, and these exploratory models can play a valuable role as part of an iterative model building process. We look forward to seeing future work by the authors, and indeed the wider systems thinking community, to further develop quantitative fast fashion models that further illuminate this important fixes that fail dynamic. Open access funding provided by IReL
Biology · Business · Discards · Fishery · Fishing · Obsolescence · Computer Science · Consumer Perception and Purchasing Behavior · Cultural and Historical Studies · Fashion and Cultural Textiles · Marketing
| Citations per year | 1 |
|---|---|
| Citation span | 2026 - 2026 (1) |
| Highly cited | No |