Cutting resistance assessment for three varieties of cassava roots

Abstract

Confronted with limited data on the physical and mechanical properties of cassava peel and root to aid machine developers, an efficient model has been developed to assess the peel and root resistance to cutting. The evaluation was done with knife thicknesses of 1.0, 1.5 and 2.0 mm using cassava varieties such as Duade kpakpa, Dudze and Sika bankye as experimental samples for three postharvest delays. The knife penetrated the tuber at 50, 100 and 150 mm away from the proximal end. An average peel thickness obtained was in the range of 1.81 – 3.01 mm. The average diameters recorded ranged from 52.52 mm to 60.40 mm. The cutting resistance assessed for the Duade kpakpa, Sika bankye and Dudze cassava roots were 44.85 N, 50.01N and 53.53 N respectively with no significant differences (P ˃ 0.05). Furthermore, the strength of the peel for the Sika bankye root was 10.31 % higher than that of the Duade kpakpa root and the resistance of the Dudze peel was 6.58 % higher than that of the Sika bankye root. The penetration resistance of the tuber decreased with increasing postharvest delay and increased with increasing knife thickness. The effect of the treatments on the response variables evaluated by factorial analysis showed that significant differences generally decrease with increasing interation. Comparing the results, cutting resistances of 229.02 N, 223.09 N and 204.43 N in maximum were obtained for the Dudze, Sika bankye and Duade kpakpa cassava roots, respectively. The average cutting resistance of Dudze cassava root was 0.24 % higher than the value for Sika bankye root and the value for the Sika bankye root was 7.82 % higher than the Duade kpakpa cassava roots. It is therefore clear that the penetration resistance of the cassava cultivar to compressive cutting could be ranked as (Dudze cassava root > Sika bankye root > Duade kpakpa cassava root). The quantitative assessment by PLSR model under the knife thicknesses (R² = 0.9689; RMSE = 2.1020) was significantly better than PLSR model under postharvest delay (R² = 0.7845; RMSE = 4.0183). The technique employed in assessing the cutting resistance gave emphasis to the cultivar differences and provided a measuring sequence and outstanding quantitative analysis.