News:

Simutrans.com Portal
Our Simutrans site. You can find everything about Simutrans from here.

Advantage to 'stockpiling' freight?

Started by AP, January 30, 2014, 09:27:41 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

AP

Just wondering, do industrial chains still stop producing if the destination receives too much supply at once?

If so, is there a gameplay advantage to the player, if using large ships to transport significant amounts of cargo, to transporting it 99% of the way and then trickling it the final few yards in a steady manner (e.g. by horse and cart) rather than delivering say 1200 tonnes of coal all at once?

I mean, does it encourage the supply industry to work in a more reliable fashion (easier to balance routes etc?). Obviously it would be an extra layer of complexity to introduce...

jamespetts

Demand will indeed cease if the end consumer has enough stock; but there is also a "maximum intransit percentage", which limits the number of goods that can be in transit to the destination at any one time, and, if this is exceeded, demand will also stop. This figure is based on the ratio of the time that it takes the end consumer industry to empty its store at maximum consumption rate to the average time that it takes goods to arrive.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

Sarlock

I am doing a test game and I can't achieve anything near 100% utilization in the end factory.  The problem is that there are two conditions for calling freight from the supplier: sufficient capacity in storage and sufficient capacity in the inbound network of goods.  If either of these conditions is not met, more cargo is not produced and the transportation network sits idle.  As soon as the end consumer gets below its stockpile level, a burst of activity happens until either enough goods in transit hits the limit or another delivery arrives and overloads the stockpile at the consumer again.

As I indicated in my other post, I think the easiest solution to this issue is to significantly raise storage capacity at the consumers (or even remove the limit completely).  This allows "in transit" to be the primary controlling mechanism through which freight is throttled so that a well planned network can achieve 100% factory utilization.  This smooths out the demand curve so that as the end consumer uses up 10 units of cargo, another 10 is demanded on the network (or, a total transit time change is made through a more/less fast network).  Now we get 10 added to supply every time 10 is consumed rather than these bursts of new cargo every time the stockpile is below storage limits.

Example:

Coal merchant: 70 storage
Network can hold 400 maximum cargo based on avg transfer time

Boat leaves colliery with 400 units cargo and reaches coal merchant 1 month later.  Coal merchant consumes this stock while the boat returns to the colliery.  While the coal merchant is consuming this inventory, the colliery and its dockyard are not receiving any new cargo.  Once the coal merchant is below 70 units, it will then call for more cargo, spurring the colliery in to action to produce a bunch more coal.  This coal is loaded on to the returning boat and zips off back to the coal merchant.  Meanwhile, the coal merchant runs out of coal and sits idle for half a month or more until the next shipment arrives.
Current projects: Pak128 Trees, blender graphics

jamespetts

Increasing storage capacities is a realtively easy thing to do in principle, although consideration will have to be given to the amount by which they should be increased (which will be different for different types of industries and in different eras, so much detailed consideration will be necessary). Removing the limit entirely would require quite drastic re-working of the code, including some quite fundamental reconceptualisation, not least because the maximum intransit percentage is a percentage of the very limit that you suggest abolishing, so this number would have to come from somewhere else.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

Sarlock

I understand... I didn't realize the two-way connectiveness of this number.  I wonder if you could just use monthly production/consumption instead.  Storage is set around 20% +/- of production rates... I wonder if you'd achieve a similar result for "in transit" limitations by using a fraction of production level.  The only complication I can see is that this number is dynamic due to passenger/mail/electricity supply.  But on the flip side, this would also help scale up "in transit" amounts for a well supplied industry, whereas right now that factor doesn't affect storage.
Current projects: Pak128 Trees, blender graphics

MCollett

Quote from: jamespetts on January 31, 2014, 12:12:12 AM
the maximum intransit percentage is a percentage of the very limit that you suggest abolishing, so this number would have to come from somewhere else.

But that cancels out.  What matters in the end is not the maximum intransit percentage, but the maximum intransit amount, which is calculable directly from the consumption rate and the lead time for resupply, independently of the storage capacity.  If I understand correctly, the current code first divides the amount by the storage capacity to get the percentage ... and then multiplies by the capacity again to get back to the amount!  So when transit capacities are being determined dynamically, the whole idea of 'maximum intransit percentage' is redundant, regardless of whether maximum storage capacities are themselves retained, modified or abolished.

Best wishes,
Matthew

AP

#6
On a related matter - warehouse maintenance costs. If warehouses/ as part of stations are an essential part of freight management, the maintenance charges need carefully considering. I'm puzzled for instance why there is a fairly steep (in 1750) monthly maintenance charge for a heap of coal sitting on the ground.

It's cheaper to build 10 staging inns than a warehouse, you get more storage and lower monthly maintenance.