This post is part of a 4-part series on Capacity Planning.
Planning long-term capacity is the first step in capacity planning. As mentioned previously, we won’t dig too much in sales forecasting and demand planning topics, but clearly appropriate long-term forecasts and strategies are needed to plan for major pieces of equipment accordingly. When an organization has historical sales data for its products (or for similar products when introducing new ones), it is often a very good starting point. Various quantitative techniques can project current numbers forward in time. For example, time series forecasting, with moving averages and weighted moving averages, can be of interest for short and possibly mid-term forecasting – but can be a bit more hazardous on longer terms. Another interesting way to move forward in time, perhaps for more extended periods, is to find, when possible, one or many external parameters that can correlate with our sales. For example, demography will drive demand for many products everyone regularly consumes that cannot be reused (ex: food, soap, energy), and household income trends can drive demand for luxury products. In the end, various experts can also chime in to weigh the impact of miscellaneous other elements, such as new technologies introduction or changes in legislation.
Finally, we should always keep in mind that forecasts are inaccurate by nature and use them wisely. However, their accuracy tends to increase when we look at aggregated data and for shorter periods.
Make-to-stock (MTS)
Example
I’m no expert on toilet paper manufacturing, so please be indulgent when I start throwing random numbers in this example to illustrate the manufacturing concepts discussed.
Let’s start throwing some numbers in our toilet paper example now. This fictitious organization’s sales forecast is mainly based on historical data and analyzing demography. When 2020 began, the organization believed that it could sell 32,000,000 kg of toilet paper this year (and expected to have sales equally distributed throughout the year). Since demography is on an upward trend, and the organization is a competitive player in its markets, the organization believes that in the longer term it will be able to sell an additional 1,000,000 kg of toilet paper every year on average for the next 10 years, just like it did for the last 5 years. When bringing back 32,000,000 kg/year daily (considering a seven-day work week), it needs to produce roughly 88,000 kg per day this year. And it will need to be able to increment this production by 2,800 kg each year until reaching a total output of 116,000 kg/day in 10 years from now.
Year | Yearly Demand (kg) | Daily Demand (kg) |
2020 | 32,000,000 | 88,000 |
2021 | 33,000,000 | 90,800 |
2022 | 34,000,000 | 93,600 |
2023 | 35,000,000 | 96,200 |
2024 | 36,000,000 | 99,000 |
2025 | 37,000,000 | 101,800 |
2026 | 38,000,000 | 104,600 |
2027 | 39,000,000 | 107,200 |
2028 | 40,000,000 | 110,000 |
2029 | 41,000,000 | 112,800 |
2030 | 42,000,000 | 114,600 |
2031 | 43,000,000 | 116,000 |
Table 1: Yearly and daily expected demand for the next 10 years
This sales forecast is critical for long-term capacity planning and to ensure the plant’s installations will be able to meet customer demands competitively: there is some room to grow within existing installations by little optimizations here and there, but typically one can hardly double production without significant investment on the shop floor. With the original production line, up to 25,000 kg of toilet paper could be manufactured per 6h work shift, including maintenance, but not changeovers. The organization then had a maximum theoretical capacity of 100,000 kg/ day. But for a single type of roll (no changeovers), if it managed to staff weekends and night shifts properly, and when no incident stops production! It had to invest to continue growing. So they decided to renew their installations with new lines B and C that can now process 30,000 kg per 6h work shift (15,000 kg each). They could now theoretically produce up to 120,000 kg/day.
Line | Daily Capacity (kg) |
B (multi-ply abilities) | 60,000 |
C (various quilting patterns) | 60,000 |
D (reengineered line A) | 100,000 |
Table 2: Daily production capacity per production line
Notice here I no longer talk about single products (2 or 3-ply, quilt patterns, prints). I’m not even considering the number of rolls – the thinner it is, the more we use anyway! This helps increase the forecast’s accuracy: we can’t be sure exactly how people’s preferences on toilet paper will evolve, but they will likely continue to wipe themselves with a similar amount of toilet paper. Unless there is a big breakthrough with bidets or reusable toilet paper.
Up next in the series will be blog posts on: