Remanufacturing Simulation

This is the final post in a three-part series arising from the talk that I did with Dr Joe on the subject of remanufacturing. (See also: Motorcycle upcycle, and Recycling: the Elephant in the Room.)

We’ve established that recycling isn’t sufficient… which means we’re going to have to get more utility out of things when they reach the end of their normal life. Perhaps remanufacturing holds the key?

Trouble is… compared to the factory, showroom, or retail environment… things at the end of life are messy. At the end of life, products get treated in unusual ways. Old phones get given to relatives, or left in a drawer; old appliances get taken to the rubbish tip, or pushed into a sinkhole. It seems that old computers are often kept because people aren’t confident about erasing all the personal information they contain. Old cars… they provide a good illustration of the extent of the problem.

You could drive a new car out of the showroom, and wreck it fifteen minutes later. Another owner might avoid accidents, and choose to keep the same model for fifteen years or more. One owner might live by the ocean, and see their pride and joy rust badly enough to ruin it in five years, while another lives in the desert and experiences no corrosion at all (although the paintwork fades, affecting its value quite quickly). Some people like “hard driving” while others seek fuel economy. Some take care to have the vehicle checked out at every servicing interval, while others treat the machine with less respect. Taxi drivers cruise around town all day, while salesmen eat up motorway miles…

Every one of those vehicles is a different prospect at the end of life, and they reach that point for different reasons. The parts that are salvageable, and the appropriate remediation strategies are completely different in each case – as are the timings involved.

Guide [1] summarised the problems as “stochastic product returns, imbalances in return and demand rates, and the unknown condition of returned products.”

What does this mean? It means if you’re a manufacturer and you’re hoping to get things back so that you can recondition them, you don’t know how long you’re going to have to wait. When end-of-life products start coming back, there’s no guarantee that people will still be interested in them when you’ve reconditioned them… and what you get back might be spoilt beyond any possibility of economic repair.

Compare that to the simple, sterile world in which you dig materials out of the ground, and turn them into all-original products as quickly as possible.

Let’s imagine that you do have a product where remanufacturing is appropriate. There are some success stories. For example, most photocopier companies don’t actually sell photocopiers anymore: they lease them, which means that they can expect to get them back at end-of-life. (Why does this work so well for photocopiers? Because the size and shape of a piece of paper isn’t going to change any time soon: the ‘guts’ of a photocopier can be salvaged, checked over and made to serve again, in an improved model with a better user interface, but the same basic core.

Kodak disposable camera

A special case: at the peak, some 92% of the ‘single use camera’ was reused… but obviously, the customer has to return this product, or they never get their pictures.

Just imagine, though, that you’re running a factory that supplies a product, and you’re hoping to do some remanufacturing. What does that mean? Well, at the beginning, you’re not doing much remanufacturing at all, because none of your products have reached the end-of-life phase… so for a while, you’re a pure manufacturer. Later, used products begin to come back, and some useful parts can be saved, and made to serve again.

That means you’re going to need to retrain manufacturing staff to perform a disassembling and checking role. You’re going to need space to store returned products, and reclaimed parts… and you’re going to need to vary the volume in which you buy or make new parts, to take into account the volume of reclaimed parts that you’re getting back… which will vary each month. If you’ve ever heard of an economic order quantity, or economic batch sizing (or if your Enterprise Resource Planning system is predicated upon those concepts) you will recognise the pitfalls that are posed by the return of an unknown quantity of parts at an unknown time, in unknown condition.

Remanufacturing is messy, in every sense of the word.

What was needed was a tool for understanding the complex nature of end-of-life scenarios, and that was the purpose of the simulation model that I originally described at the International Conference on Remanufacturing in Glasgow, back in 2011 [2]. The theory goes something like this: you use a simulation language that’s designed for constructing manufacturing simulations, but instead of simply having workpieces flowing through a factory, being assembled into a product and then quitting the system as finished goods, you send them into a loop, where they continue to circle around a ‘use phase’ submodel.

Every month, the product gets some ‘wear and tear’ assigned to its component parts, and becomes an end-of-life product if the accumulated wear is enough to break it. It’s also checked against a specified chance that it suffers an accident, and there’s a chance that products over a certain age are found to be unwanted, and leave the ‘use phase’ loop. Products then move into an end-of-life submodel where there is a specified probability that they are returned to the manufacturer. If they are, they get disassembled, and the condition of components is assessed. Those that are good enough to serve again go into the queue for component assembly, and cause a corresponding reduction in the orders for brand new components.

Basically, it’s a machine for telling you how much remanufacturing you can expect to be doing, and when… based upon the conditions that you specify. For example, you could explore a leasing-based business model (like the photocopiers) by perhaps specifying a 48-month useful life, after which there is a 90% chance that the product is immediately retired (because it’s specified in the conditions in the leasing contract) with a 100% chance that the product comes back for remanufacturing.

Model inputs spreadsheet

This spreadsheet allows the user to specify whatever model conditions they want to explore – without having to edit the model itself.

Alternatively, if the product that you’re interested in was the engine in a tank, you might say that there’s a 2% chance of accidental damage every period, a fixed 60-month service life, and an 90% chance of the engine coming back for remanufacturing when it reaches end-of-life. (Some are ruined on battlefields, get abandoned, or end up in museums, and so on…)

The result of all this modelling comes in graph form, exported by the simulation tool, to show how much remanufacturing we can expect to be doing. Back in 2007, Geyer et al [3] had predicted that the volume of remanufactured products released over time would be a trapezoidal subset of the whole, and our simulation anticipates the same result, only a little bit messier due to the randomness introduced in an effort to represent the uncertainties of real-world conditions.

Graph showing proportion of remanufacturing occurring

Predicted quantity of reused components, based upon a scenario defined within the model. It’s… kind of trapezoidal!

It suggests that you’ll only ever be able to do some remanufacturing, and will still need a source of virgin components while volume manufacturing continues. On its own, then, remanufacturing doesn’t ‘close the loop’ – and it doesn’t ‘save the planet’, although it helps out a bit. It slows the rate at which we consume resources… at the cost of additional complexity in the supply chain, and in the scheduling of operations.

(The slides from our talk can be seen on



  1. Guide, V.D.R. (2000) Production planning and control for remanufacturing: industry practice and research needs, Journal of Operations Management, Volume 18, Number 4, June 2000, pp. 467–483
  2. Farr, R. and Lohse, N. (2011) Use of Enterprise Simulation to Assess the Impacts of Remanufacturing Operations, Proceedings of the International Conference on Remanufacturing, ICoR 2011, 27–29 July 2011, University of Strathclyde, Glasgow. (Link to full paper here.)
  3. Geyer, R., Van Wassenhove, L.N. and Atasu, A. (2007) The Economics of Remanufacturing Under Limited Component Durability and Finite Product Life Cycles, Management Science, 53(1), pp. 88–100

One thought on “Remanufacturing Simulation

  1. Pingback: Greener Warehouses… greener still | Capacify

Talk to me! Leave a reply...

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s