Process Engineering Associate – Pre interview questions

Waste Conversion LTD is a company that treats organic waste using worms to form organic fertilizer and an insect based animal feed. In this simplified process, different types of organic waste (cabbages, bananas, tomatoes, manure…etc) are delivered to the factory using trucks. As a pre-processing step, an industrial waste mixer is used to mix the waste at different predetermined compositions to form several recipe batches. These waste recipe batches are A, B, C and E.

Depending on the production plan, a predetermined number of beds (with unique identifiers) are loaded with the batches of waste at different recipes. Typically, one bed is loaded with one mixed waste batch that weighs 1500 kilograms. Thereafter, young/small worms at either of two loading densities (17.8 kg or 32 kg) are added to the beds and left to feed on the waste for a specified amount of days. After the days have passed, the grown/adult worms are harvested from the beds and weighed – from which growth factors are calculated. The remaining waste/compost is routed to the next processing step.

Please use the data sheet provided (excel workbook) labelled “Data sheet” to answer the questions below.

1. Please review the database (excel workbook) provided and do the following: a. Using measures of central tendencies, develop a dashboard on MS Excel or

Google Sheets showing both weekly and monthly production performance at loading densities of 17.8 kgs and 32 kgs with regard to: (1) Growth factors and (2) Feed yields. In this analysis, clearly distinguish between production beds with a cycle time below 17 days and those with cycle times above 18 days.

b. Improve on the above by developing a control chart displaying daily production performance with regard to growth factors for beds with cycle times below 11 days at loading densities of 17.8 kgs. Include upper and lower control limits to the plots.

2. Typically, waste mixed with the recipe A composition is used owing to higher historical yields. However, during the production period (Jan to October 2018), it was noted that the worm growth factor varied significantly when recipes B, C and E were used at different loading densities (17.8 kg and 32 kg). As the quality engineer, please use analysis of variance or other known statistical methods to answer the following pending questions (using the data in the spreadsheet):

a. Is it best to use recipe A for all beds at any loading density? b. In the event that recipe A is not available, what recipes (between B, C and E) do

you recommend using and at what loading density? Please give these in descending order based on growth factors.

Please note that the data is not clean i.e. there are outliers as well as missing data points. All definitions have been provided in the data sheet (excel workbook)

https://docs.google.com/spreadsheets/d/1ruu9g8tyZh7XSYspbSrSncNV-tRM-7iyYNtkN7stl_k/edit?usp=sharing
Please provide your answers/responses in the MS Excel with referenced calculations/analyses where necessary. If successful, note that you will be required to interpret your responses in detail.