CHANGELOG.md

Changelog

All significant changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

[1.0.1]

Added

• Option to suppress dummy-product indices in simulation output.
• SupplyChainNetwork.nodes_by_index dict, which returns a SupplyChainNode object for a given index. Use this in place of SupplyChainNetwork.get_node_from_index(), which will be deprecated.

Changed

• Various speedups.
• DemandSource.lead_time_demand_distribution() now returns an rv_discrete object with a single value (0) and a single probability (1) if the lead time is 0. (In that case, the lead time demand always = 0.)
• Updated dependency for setuptools package to require v70.0 or later, to avoid security vulnerability.

Fixed

• Bug that caused mwor_system() to crash when demand_source is provided as argument #146.
• Bugs that caused improper handling of orders and shipments when multiple suppliers provide the same raw material. #171
• Bug that caused ssm_serial.optimize_base_stock_levels() to crash when all lead times are 0. [#173]

[1.0.0] -- 2024-05-13

Introducing: Products

• Stockpyl now supports products in simulations. Products are implemented using the supply_chain_product object.
• Products are "handled" by nodes. Most attributes (stockout_cost, inventory_policy, etc.) may be specified either at the node level, the product level, or the (node, product) level.
• Products are related to each other via a bill of materials (BOM), which specifies the number of units of an upstream product (raw material) that are required to make one unit of a downstream product (finished goods).
• Only the simulation features of Stockpyl can currently make use of products; MEIO and other features still assume a single-product model.
• For more information about creating and managing products, and simulating multi-product systems in |sp|, see the supply_chain_product module or the tutorial page for multi-product simulation.

Added

• Products.
• helpers.nearest_dict_value() function, to find key in a dictionary that's nearest to a given number and return the corresponding value.
• loss_functions.standard_normal_loss_dict() function, to build a dictionary of loss-function values.
• Support for negative binomial demand distributions in DemandSource.
• Functions to validate and parse nodes and products within a network or node. (Mostly used internally for simulation.)

Changed

• Requires Python 3.8 or later.

• Default git branch has been changed to main. If you have a local clone, you can rename it using:

   git branch -m master main
   git fetch origin
   git branch -u origin/main main
   git remote set-head origin -a
  

• NodeStateVars.raw_material_inventory is now indexed by product, not by predecessor.

• NodeStateVars object is now in its own module, node_state_vars.py, rather than in supply_chain_node.py.

• NodeStateVars attributes should now be accessed using the appropriate methods, rather than using the attribute directly. For example: use node.state_vars[42].get_outbound_shipment(successor, product) instead of node.state_vars[42].outbound_shipment[successor][product]. These methods allow you to omit product info if it is inferrable. (This is especially useful for models that don't have products explicitly added.)

• More compact text representation of SupplyChainNetwork and SupplyChainNode objects via __repr__().

Fixed

• supply_chain_network.network_from_edges() now only creates a DemandSource for sink nodes or if the demand source parameters were provided specifically for that node in the input args.
• Bug in helpers.ensure_list_for_time_period() that caused it to handle numpy arrays improperly.
• Bug in demand_source.py that sometimes caused infinite recursion when some attributes were None.
• Bug in sim_io.py that caused incorrect headers for a few state variables.
• Various other bug fixes.

Known Issues

(See https://github.com/LarrySnyder/stockpyl/issues for all issues.)

• BOM relationships assume the products are infinitely divisible; e.g., if 5 units of product A are required to make 1 unit of product B, and there are 4 units of product A available, then 0.8 units of product B are produced. #155
• Lead times are specific to downstream node and finished good product, not upstream node and raw material. This makes it hard to have different lead times for ordering different raw materials from different upstream nodes. #149
• Echelon base-stock policies are not working reliably yet, at least for systems with multiple products. #153
• Disruptions can only occur at node level, not product level. #158

[0.0.15] -- 2024-02-10

Added

• Tutorial, in Jupyter notebook form, to accompany Snyder, L. V., "Stockpyl: A Python Package for Inventory Optimization and Simulation," in: Bish, E. K. and H. Balasubramanian, INFORMS TutORials in Operations Research, 156–197, 2023
• Check to make sure that a node's inventory_policy attribute is set, and that the Policy object's node attribute is set to the node
• Set the Policy object's node attribute automatically when setting the node's inventory_policy to it
• Ability to print simulation output table to text file (in addition to CSV option)
• order_capacity attribute for SupplyChainNode, which is respected in simulation

Changed

• Changes to optimize_base_stock_levels() in ssm_serial that result in large speedups in some cases (via @DominikKamp)
• Add vertical line between nodes in simulation output to make it easier to separate nodes visually
• Minor edits to documentation and README

Fixed

• Incorrect handling of custom demand distributions in ss.s_s_discrete_exact()
• Error in unit tests for meio_general.truncate_and_discretize()

[0.0.14] -- 2022-08-17

Fixed

• Bug in feature that allows overriding order quantities in simulation

[0.0.13] -- 2022-08-10

Added

• Features to run simulation period-by-period instead of all at once (useful for using in a reinforcement learning (RL) context)
• newsvendor.newsvendor_poisson_explicit() function added, for explicit-form newsvendor problem with Poisson demands
• Various unit tests
• More robust attribute handling for various objects

Changed

• Allow earlier version of numpy (v1.21 insead of v1.22), which in turn allows Python 3.7 instead of 3.8
• New format for storing and loading named instances, and better documentation
• newsvendor.newsvendor_explicit() parameter renamed to revenue from selling_revenue for consistency

Fixed

• State variables are stored as int and float instead of int64 and float64 so they can be JSON serialized when saving
• Bug that created spurious entries in inbound_order_pipeline
• Bug in which only order LT was respected in simulation if both order and shipment LTs are nonzero
• Bug in default value for consistency_checks in simulation()

[0.0.12] -- 2022-07-12

Changed

• Looser dependencies (previous versions required newer than necessary versions of some packages)

[0.0.11] -- 2022-07-11 -- YANKED

[0.0.10] -- 2022-07-11 -- YANKED

[0.0.9] -- 2022-06-28

Added

• Methods to convert objects to and from dicts

Changed

• Removed dependency on pandas

Fixed

• Bug that prevented named instances from being loaded correctly using load_instance()

[0.0.8] -- 2022-06-27

Changed

• Removed cost summary output and print_cost_summary option from sim_io.write_results()

[0.0.7] -- 2022-06-27

Fixed

• Small issue in README file

[0.0.6] -- 2022-06-27

Added

• Various new unit tests

Fixed

• Bug in supply_chain_network.echelon_to_local_base_stock_levels()

[0.0.5] -- 2022-06-25

Added

• Named instances from Shang and Song (2003)
• More flexible options for node indexing in ssm_serial.optimize_base_stock_levels() and ssm_serial.newsvendor_heuristic()
• More flexible options for choosing rows and columns in simulation output in write_results()
• Better handling of disrupted items for type-RP disruptions
• Various new unit tests

Changed

• Smarter initialization of inventory levels in simulation
• ssm_serial.optimize_base_stock_levels() recognizes if demand distribution is discrete and handles discretization accordingly
• Default disruption type changed to 'OP'

Fixed

• Bug in order quantity with type-RP disruptions

[0.0.4] -- 2022-06-14

Added

• Various new unit tests
• Ability to pass kwargs to SupplyChainNetwork()
• owmr_system() function to build OWMR networks

Changed

• Improved flexibility of inputs in network_from_edges()

Fixed

• Bug in which inbound_order_pipeline was initialized incorrectly in simulation

[0.0.3] -- 2022-06-11

Added

• Various new unit tests.
• Improved documentation throughout.
• Feature to check for duplicate edges when adding edges to a network.
• Feature to check for directed cycles before running simulation.

Fixed

• Various bugs in backorder calculations in simulation, especially with disruptions.

[0.0.2] -- 2022-06-06

Fixed

• Fix problems with setup that prevented installation via pip.

[0.0.1] -- 2022-06-02

Initial release.