The unit of analysis is the garment workshop. We measured the productive efficiency of each of the 15 participating workshops from the existing cluster in the City of Yaguarón. 

Starting with a pool of 24 interested workshops, 18 agreed to participate of the experiment. We assigned garment workshops to the treatment group through random sampling of 9 blocks of garment workshop pairs. Each block included 2 garment workshops with the greatest possible similarity, based on the following variables, which we collected before the intervention: 1. Number of sewing machines per worker 2. Number of specialized machines per garment workshop 3. Membership of Yaguaron’s Garment Workshops Association 4. Type of electricity connection (single-phase or three-phase) 5. Number of collaborators (women and men) On each of the 9 blocks, one garment workshop was randomly chosen and assigned to the treatment group, while the other was assigned to the control group. As a result, both treatment and control groups were balanced with respect to the five control variables we described previously. After randomization, 3 garment workshops dropped out of the study. The experiment was finally carried out with 9 garment workshops in the control group and 6 garment workshops in the treatment group. The balance was not greatly affected by this change in the sample and both groups continued to keep a high degree of similarity. In general terms, the intervention sought to improve the productive efficiency of garment workshops during the production of a new product. For this investigation, productive efficiency refers to the efficiency achieved during production, measured in minutes worked per product and per workshop, also considering the number of garments reprocessed and the number of garments rejected during the quality control. To calculate the minutes worked in the production of one school overall (𝑀𝑇) in the workshops of Yaguarón, we measure the following variables during the production phase:  1. Basic minutes (𝑴𝑩): is the actual time taken to complete the production of one school overall in each garment workshop. 2. Standard allocations per package produced (𝑨𝑬𝑷): it is the sum of 10% of the value of basic minutes achieved by each workshop, which is based on standardized metrics that affect performance during the productive process. 3. Assignments for machinery and operators (𝑨𝑴𝒚𝑶): it is the sum of 20% of the value of basic minutes achieved by each workshop, based on standardized performance and productivity metrics, understanding that the productive of each day is not constant and that there are low, regular, and high productivity levels depending on external factors. Thus, the minutes worked in producing one school overall (𝑀𝑇), are calculated as follows: 𝑀𝑇 =𝑀𝐵 +𝐴𝐸𝑃 +𝐴𝑀𝑦𝑂 (1) To estimate the productive efficiency of the workshops, we also resort to another variable used in the literature to measure production processes. The standard minute allowed (SAM) is the time that is allocated to a task or work content in a garment in a given industry (Abtew, et al., 2019). In other words, it is the time it takes an operator to complete the corresponding task within the productive process indicated in the operational sequence, with 100% of efficiency. In our case, the SAM=30.77 minutes is a constant that reflects the standard time (minutes of production) in which the local garment industry produces a unit of school overalls. Using these concepts, the main variable of the study is the productive efficiency measured as a proportion of minutes worked per batch in relation to the SAM. The formula used is the following: Productive efficiency=MT/SAM×100 It is important to mention that as production is faster (in minutes) and the quality is maintained (rejections do not increase), efficiency is gained. In this sense, a lower indicator value represents a gain in efficiency. In order to calculate the indicators for the treatment and control groups, we register the information needed for each garment industry, using a survey that was completed by the instructors. 

3 months

Approximately 20% of the time of our Head of Experimentation, 20% of the time of our Head of Solutions Mapping, 20% of the time of our Head of Exploration, plus 50% of the time of two people hired for this learning cycle was dedicated to the design, management, analysis, reporting, and monitoring of results, throughout the duration of the intervention.  Moreover, the lab also counted with the help of, at the time, a Data Analyst, who devoted over 50% of her time to both implementation and analysis of the experiment. And for a brief time at the beginning, we had a part-time assistant helping in the process of sending out invitations to workers.  

We observed the following limitations that could be improved:  1. The limited number of the sample: with a larger sample, more robust results could be obtained, covering more types/profiles of garment workshops. Likewise, a greater sample may allow results to be extrapolated to a wider variety of workshops. 2. The product that the workshops made for the experiment is not the product in which they are specialized: due to a UNDP policy, the overalls made by the workshops were donated to a school. Therefore, school overalls had to be made, instead of jeans, which is the product in which the Yaguarón clothing cluster is specialized. Although the workshops responded with success, the training and results could be more useful to the community if they were based on the product they manufacture on a daily basis 3. In the data collection, we considered only the information necessary to create the productive efficiency indicators. In another iteration of the experiment, other relevant variables could be collected, with the aim of not only calculating the degree of productive efficiency of the workshops, but also exploring its drivers.

9

For evaluation purposes only:  1. Payment to the garment workshops for the school overalls made~ $24000 2. Other resources: time invested by AccLab Members (average or 2h-4h/week over the full period of the pilot), half time engagement of two specialists hired for this learning loop. Time invested by the SNPP instructors. 

Yes, from this activity, we detected the need to have more information on the garment workshops of the Yaguarón cluster in terms of infrastructure, investment, productivity, employment, formalization, among others, for which we designed and implemented the First Census of Workshops of clothing in Yaguarón in 2022. Moreover, our new Poverty Innovation Funding Window Project will adapt this intervention, integrating some elements of the In Motion UNDP Program (https://www.undp.org/sdgvaluechains/motion).  

Sort-of: within our new Poverty Innovation Funding Window, we are taking some of the elements of this intervention to scale, but adapting it to follow more closely the model of the In Motion UNDP Program. This will be done in 3 clusters of garment workshops: one more time in the same cluster of this experiment, another in the city of Pilar, Paraguay, and one final intervention in a cluster of San Lorenzo, Paraguay. 

No, this experiment did not require iterations.

1. Once the geographical area and the dates of the intervention have been defined, make an approach to organizations/associations or leaders that allow a more direct contact with the garment workshops.   2. Survey the training needs of the workshops before defining the intervention, ensuring (if possible) that these trainings are based on the daily activities of the workshops, so that they can really incorporate the knowledge into their own production processes and the effect of the intervention is more lasting.

Yes, this experiment can be replicated in any productive sector where the factories, workshops, and workers in general need training in quality control and industrial production, and when it is expected that an improvement of the productive efficiency can be an ally to reduce the barriers to employment formalization.

Both sensing and exploration activities influenced greatly the design of the experiment, the hypotheses we decided to test, the places and people we engaged, and the things we needed to observe as the experiment unfolded.  The activities designed as a part of the mapping and exploration phases, make it possible to made it possible to know in depth the needs of the actors involved, so that the experimentation was fully based on the findings of these phases.

The response capacity of the participating workshops, which quickly learned the production processes and the manufacturing technique of a totally new product and carried it out with high quality standards. In addition, we were surprised by the family networks and the connections between the different workshops, which, although allowed them to respond to the demand and generate constant synergies between the workshops, is a double-edged sword, since while they generate agile exchanges between workshops, they can also perpetuate their informal status over time. And finally, it was surprising how open and collaborative were both the Education and Science Ministry and the Private Garments Company we engaged for this experiment. In a very short period of time, both these institutions moved fast to make this experiment happen.